Federal Register/Vol. 85, No. 100/Friday, May 22, 2020/Rules and Regulations

31286 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations

ENVIRONMENTAL PROTECTION documents in the docket are listed on reference purposes, the EPA defines the AGENCY the https://www.regulations.gov/ following terms and acronyms here: website. Although listed, some CAA Clean Air Act 40 CFR Part 63 information is not publicly available, CAMR Clean Air Mercury Rule [EPA–HQ–OAR–2018–0794; FRL–10008–60– e.g., confidential business information CEMS continuous emissions monitoring OAR] or other information whose disclosure is systems restricted by statute. Certain other CFR Code of Federal Regulations RIN 2060–AT99 material, such as copyrighted material, CRA Congressional Review Act is not placed on the internet and will be EGU electric utility steam generating unit National Emission Standards for EPA Environmental Protection Agency Hazardous Air Pollutants: Coal- and publicly available only in hard copy EPRI Electric Power Research Institute Oil-Fired Electric Utility Steam form. Publicly available docket HAP hazardous air pollutant(s) Generating Units—Reconsideration of materials are available either HCl hydrochloric acid Supplemental Finding and Residual electronically through https:// HF hydrogen fluoride Risk and Technology Review www.regulations.gov/, or in hard copy at HQ hazard quotient the EPA Docket Center, WJC West ICR information collection request AGENCY: Environmental Protection Building, Room Number 3334, 1301 km kilometer Agency (EPA). Constitution Ave. NW, Washington, DC. MACT maximum achievable control technology ACTION: Final rule. The Public Reading Room hours of MATS Mercury and Air Toxics Standards operation are 8:30 a.m. to 4:30 p.m. MIR maximum individual risk SUMMARY: The U.S. Environmental Eastern Standard Time (EST), Monday Protection Agency (EPA) is revising its MW megawatt through Friday. The telephone number NAAQS National Ambient Air Quality response to the U.S. Supreme Court for the Public Reading Room is (202) Standards decision in Michigan v. EPA, which 566–1744, and the telephone number for NAICS North American Industry held that the EPA erred by not the Docket Center is (202) 566–1742. Classification System considering cost in its determination NEI National Emissions Inventory that regulation under section 112 of the FOR FURTHER INFORMATION CONTACT: For NESHAP national emission standards for Clean Air Act (CAA) of hazardous air questions about these final actions, hazardous air pollutants pollutant (HAP) emissions from coal- contact Mary Johnson, Sector Policies NOAEL no-observed-adverse-effect-level and oil-fired electric utility steam and Programs Division (D243–01), NOX nitrogen oxides generating units (EGUs) is appropriate Office of Air Quality Planning and NTTAA National Technology Transfer and Standards, U.S. Environmental Advancement Act and necessary. After primarily OAQPS Office of Air Quality Planning and comparing the cost of compliance Protection Agency, Research Triangle Standards relative to the benefits of HAP emission Park, North Carolina 27711; telephone OMB Office of Management and Budget reduction from regulation, the EPA number: (919) 541–5025; and email PB–HAP hazardous air pollutants known to finds that it is not ‘‘appropriate and address: [email protected]. For be persistent and bio-accumulative in the necessary’’ to regulate HAP emissions specific information regarding the risk environment from coal- and oil-fired EGUs, thereby modeling methodology, contact Mark PDF Portable Document Format reversing the Agency’s previous Morris, Health and Environmental PM particulate matter conclusion under CAA section Impacts Division (C539–02), Office of PM2.5 fine particulate matter POM polycyclic organic matter 112(n)(1)(A) and correcting flaws in the Air Quality Planning and Standards, PRA Paperwork Reduction Act Agency’s prior response to Michigan v. U.S. Environmental Protection Agency, RDL representative detection level EPA. We further find that finalizing this Research Triangle Park, North Carolina REL reference exposure level new response to Michigan v. EPA will 27711; telephone number: (919) 541– RFA Regulatory Flexibility Act not remove the Coal- and Oil-Fired EGU 5416; and email address: morris.mark@ RIA regulatory impact analysis source category from the CAA section epa.gov. For information about the RTR residual risk and technology review 112(c) list of sources that must be applicability of the NESHAP to a SO2 sulfur dioxide regulated under CAA section 112(d) and particular entity, contact your EPA TOSHI target organ-specific hazard index will not affect the existing CAA section Regional representative as listed in 40 tpy tons per year UMRA Unfunded Mandates Reform Act 112(d) emissions standards that regulate CFR 63.13 (General Provisions). Background information. With this HAP emissions from coal- and oil-fired SUPPLEMENTARY INFORMATION: EGUs. The EPA is also finalizing the action, the EPA is, after review and residual risk and technology review Preamble acronyms and consideration of public comments, (RTR) conducted for the Coal- and Oil- abbreviations. We use multiple finalizing two aspects of the 2019 Fired EGU source category regulated acronyms and terms in this preamble. Proposal. On February 7, 2019, the EPA under national emission standards for While this list may not be exhaustive, to proposed to find that it is not hazardous air pollutants (NESHAP), ease the reading of this preamble and for ‘‘appropriate and necessary’’ to regulate commonly referred to as the Mercury HAP emissions from coal- and oil-fired and Air Toxics Standards (MATS). and information, in whatever form, in Docket ID EGUs, thereby reversing the Agency’s Nos. EPA–HQ–OAR–2009–0234 (National Emission Based on the results of the RTR Standards for Hazardous Air Pollutants for Coal- prior conclusion under CAA section analyses, the Agency is not and Oil-fired Electric Utility Steam Generating 112(n)(1)(A) and correcting flaws in the promulgating any revisions to the Units), EPA–HQ–OAR–2002–0056 (National Agency’s prior response to Michigan v. MATS rule. Emission Standards for Hazardous Air Pollutants EPA, 135 S. Ct. 2699 (2015). 84 FR 2670 for Utility Air Toxics; Clean Air Mercury Rule DATES: Effective May 22, 2020. (CAMR)), and Legacy Docket ID No. A–92–55 (2019 Proposal). We further proposed ADDRESSES: The EPA has established a (Electric Utility Hazardous Air Pollutant Emission that finalizing this new response to docket for these actions under Docket ID Study). See memorandum titled Incorporation by Michigan v. EPA would not remove the 1 reference of Docket Number EPA–HQ–OAR–2009– Coal- and Oil-Fired EGU source category No. EPA–HQ–OAR–2018–0794. All 0234, Docket Number EPA–HQ–OAR–2002–0056, and Docket Number A–92–55 into Docket Number from the CAA section 112(c) list of 1 As explained in a memorandum to the docket, EPA–HQ–OAR–2018–0794 (Docket ID Item No. sources that must be regulated under the docket for these actions include the documents EPA–HQ–OAR–2018–0794–0005). CAA section 112(d) and would not

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00002 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations 31287

affect the existing CAA section 112(d) D. Effects of This Reversal of the G. What analysis of children’s emissions standards that regulate HAP Supplemental Finding environmental health did we conduct? emissions from coal- and oil-fired EGUs. III. Background on the RTR Action VII. Statutory and Executive Order Reviews A. What is the statutory authority for this A. Executive Orders 12866: Regulatory In the same action, the EPA also action? proposed the results of the RTR of the Planning and Review and Executive B. What is the Coal- and Oil-Fired EGU Order 13563: Improving Regulation and NESHAP for Coal- and Oil-Fired EGUs. source category and how does the Regulatory Review In this action, we are taking final action NESHAP regulate HAP emissions from B. Executive Order 13771: Reducing with regard to these aspects of the 2019 the source category? Regulation and Controlling Regulatory Proposal.2 We summarize some of the C. What changes did we propose for the Costs more significant comments regarding Coal- and Oil-Fired EGU source category C. Paperwork Reduction Act (PRA) in our February 7, 2019, proposed rule? the proposed rule and provide our D. Regulatory Flexibility Act (RFA) IV. What is included in this final rule based E. Unfunded Mandates Reform Act responses in this preamble. A summary on results of the RTR? of all other significant comments on the A. What are the final rule amendments (UMRA) 2019 Proposal and the EPA’s responses based on the residual risk review for the F. Executive Order 13132: Federalism to those comments is available in the Coal- and Oil-Fired EGU source G. Executive Order 13175: Consultation document titled Final Supplemental category? and Coordination With Indian Tribal Governments Finding and Risk and Technology B. What are the final rule amendments H. Executive Order 13045: Protection of Review for the NESHAP for Coal- and based on the technology review for the Coal- and Oil-Fired EGU source Children From Environmental Health Oil-Fired EGUs Response to Public category? Risks and Safety Risks Comments on February 7, 2019 Proposal C. What are the effective and compliance I. Executive Order 13211: Actions (Response-to-Comment (RTC) dates of the standards? Concerning Regulations That document), in Docket ID No. EPA–HQ– V. What is the rationale for our final Significantly Affect Energy Supply, OAR–2018–0794. decisions regarding the RTR action for Distribution, or Use the Coal- and Oil-Fired EGU source J. National Technology Transfer and Organization of this document. The category? Advancement Act (NTTAA) information in this preamble is A. Residual Risk Review for the Coal- and K. Executive Order 12898: Federal Actions organized as follows: Oil-Fired EGU Source Category To Address Environmental Justice in I. General Information B. Technology Review for the Coal- and Minority Populations and Low-Income Oil-Fired EGU Source Category A. Do these actions apply to me? Populations VI. Summary of Cost, Environmental, and B. Where can I get a copy of this document L. Congressional Review Act (CRA) Economic Impacts and Additional and other related information? Analyses Conducted I. General Information C. Judicial Review and Administrative A. What are the affected facilities? Reconsideration B. What are the air quality impacts? A. Do these actions apply to me? II. Appropriate and Necessary Finding C. What are the cost impacts? A. Overview D. What are the economic impacts? Regulated entities. Categories and B. Background E. What are the benefits? entities potentially regulated by these C. EPA’s Finding Under CAA Section F. What analysis of environmental justice final actions are shown in Table 1 of 112(n)(1)(A) did we conduct? this preamble.

TABLE 1—NESHAP AND INDUSTRIAL SOURCE CATEGORIES AFFECTED BY THESE FINAL ACTIONS

NESHAP and source category NAICS 1 code

Coal- and Oil-Fired EGUs ...... 221112, 221122, 921150. North American Industry Classification System.

Table 1 of this preamble is not document will also be available on the C. Judicial Review and Administrative intended to be exhaustive, but rather to internet. Following signature by the Reconsideration provide a guide for readers regarding EPA Administrator, the EPA will post a entities likely to be affected by these copy of this document at: https:// Under CAA section 307(b)(1), judicial final actions for the source category www.epa.gov/mats/regulatory-actions- review of these final actions is available listed. To determine whether your final-mercury-and-air-toxics-standards- only by filing a petition for review in facility is affected, you should examine mats-power-plants. Following the United States Court of Appeals for the applicability criteria in the publication in the Federal Register, the the District of Columbia Circuit (D.C. appropriate NESHAP. If you have any EPA will post the Federal Register Circuit) by July 21, 2020. Under CAA questions regarding the applicability of version and key technical documents at section 307(b)(2), the requirements any aspect of this NESHAP, please this same website. established by this final rule may not be contact the appropriate person listed in Additional information regarding the challenged separately in any civil or the preceding FOR FURTHER INFORMATION RTR action is available on the RTR criminal proceedings brought by the CONTACT section of this preamble. website at https://www3.epa.gov/ttn/ EPA to enforce the requirements. atw/rrisk/rtrpg.html. This information Section 307(d)(7)(B) of the CAA B. Where can I get a copy of this includes an overview of the RTR further provides that only an objection document and other related program, links to project websites for to a rule or procedure which was raised information? the RTR source categories, and detailed with reasonable specificity during the In addition to being available in the emissions and other data we used as period for public comment (including docket, an electronic copy of this inputs to the risk assessments. any public hearing) may be raised

2 The EPA took final action on the other aspect on establishing a subcategory of certain existing emissions of acid gas HAP) on April 15, 2020, in of the 2019 Proposal (i.e., solicitation of comment EGUs firing eastern bituminous coal refuse for a separate action (85 FR 20838).

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00003 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 31288 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations

during judicial review. That section of it would not remove the Coal- and Oil- CAA section 112, the EPA shall then do the CAA also provides a mechanism for Fired EGU source category from the so. the EPA to reconsider the rule if the CAA section 112(c)(1) list, nor would it The required study, which the EPA person raising an objection can affect the existing CAA section 112(d) completed in 1998, contained an demonstrate to the Administrator that it emissions standards regulating HAP analysis of HAP emissions from EGUs, was impracticable to raise such emissions from coal- and oil-fired EGUs an assessment of the hazards and risks objection within the period for public that were promulgated on February 16, due to inhalation exposures to these comment or if the grounds for such 2012. 77 FR 9304 (2012 MATS Final emitted pollutants, and a multipathway objection arose after the period for Rule). (inhalation plus non-inhalation public comment (but within the time exposures) risk assessment for mercury In section II.B of this preamble, which specified for judicial review) and if such and a subset of other relevant HAP.4 finalizes the reversal of the 2016 objection is of central relevance to the The study indicated that mercury was Supplemental Finding, the EPA outcome of the rule. Any person seeking the HAP of greatest concern to public provides background information to make such a demonstration should health from coal- and oil-fired EGUs. regarding the previous appropriate and submit a Petition for Reconsideration to Mercury is highly toxic, persistent, and necessary findings, including the the Office of the Administrator, U.S. bioaccumulates in food chains. The affirmations in the preamble of the 2012 EPA, Room 3000, WJC South Building, study also concluded that numerous MATS Final Rule and in the 2016 1200 Pennsylvania Ave. NW, control strategies, of varying cost and Supplemental Finding. Section II.C of Washington, DC 20460, with a copy to efficiency, were available to reduce HAP this preamble describes why the 2016 both the person(s) listed in the emissions from this source category. Supplemental Finding was flawed, why preceding FOR FURTHER INFORMATION Based on this study and other available the EPA has authority to revisit that CONTACT section, and the Associate information, the EPA determined in finding now, and what the EPA is General Counsel for the Air and December 2000, pursuant to CAA finalizing as the appropriate approach to Radiation Law Office, Office of General section 112(n)(1)(A), that it was satisfy the EPA’s obligation under CAA Counsel (Mail Code 2344A), U.S. EPA, appropriate and necessary to regulate section 112(n)(1)(A) as interpreted by 1200 Pennsylvania Ave. NW, coal- and oil-fired EGUs under CAA the U.S. Supreme Court in Michigan. Washington, DC 20460. section 112 and added such units to the Finally, section II.D of this preamble CAA section 112(c) list of sources that II. Appropriate and Necessary Finding explains that the EPA’s revised must be regulated under CAA section determination that regulation of HAP A. Overview 112(d). 65 FR 79825 (December 20, emissions from EGUs under CAA 2000) (2000 Finding).5 The 2000 On June 29, 2015, the U.S. Supreme section 112 is not appropriate and Finding did not consider the cost of Court ruled in Michigan v. EPA that the necessary will not remove coal- and oil- regulating EGUs in its finding that it Agency had erred when it failed to take fired EGUs from the CAA section 112(c) was appropriate and necessary to do so. cost into account in its previous CAA list of source categories, and that the Id. at 79830. section 112(n)(1)(A) determination that previously established CAA section In 2005, the EPA revised the original it is appropriate and necessary to 112(d) standards for HAP emissions 2000 Finding and concluded that it was regulate HAP emissions from coal- and from coal- and oil-fired EGUs will neither appropriate nor necessary to oil-fired EGUs. In response to that remain in place. In this preamble, the regulate EGUs under CAA section 112. decision, the EPA finalized a EPA provides a summary of certain 70 FR 15994 (March 29, 2005) (2005 supplemental finding on April 25, 2016, significant comments received on the Revision). This action was taken that evaluated cost considerations and 2019 Proposal and the Agency’s because, at that time, the EPA concluded that the appropriate and response to those comments. The RTC concluded that the original 2000 necessary finding was still valid. 81 FR document for this action summarizes Finding lacked foundation in that it 24420 (2016 Supplemental Finding). On and responds to all other significant failed to consider: (1) The HAP February 7, 2019, the EPA proposed a comments that the EPA received. reductions that could be obtained revised response to the U.S. Supreme B. Background through implementation of CAA Court decision. 84 FR 2670 (2019 sections 110 and 111; and (2) whether Proposal). In the 2019 Proposal, after The CAA establishes a multi-step hazards to public health would still primarily comparing the cost of process for the EPA to regulate HAP exist after imposition of emission compliance relative to the benefits of emissions from EGUs. First, section reduction rules under those sections. HAP emission reduction from 112(n)(1)(A) of the CAA requires the The 2005 Revision also removed coal- regulation, the EPA proposed to find EPA to perform a study of the hazards and oil-fired EGUs from the CAA that it is not appropriate and necessary to public health reasonably anticipated section 112(c) list of source categories to to regulate HAP emissions from coal- to occur as a result of HAP emissions be regulated under CAA section 112. In and oil-fired EGUs, thereby reversing from EGUs ‘‘after imposition of the a separate but related 2005 action, the the Agency’s conclusion under CAA requirements of this chapter.’’ 3 If, after EPA also promulgated the Clean Air section 112(n)(1)(A), first made in 2000 considering the results of this study, the Mercury Rule (CAMR) which and later affirmed in 2012 and 2016. EPA determines that it is ‘‘appropriate established CAA section 111 standards Specifically, the Agency proposed that and necessary’’ to regulate EGUs under of performance for mercury emissions the 2016 Supplemental Finding from EGUs. 70 FR 28605 (May 18, 2005). considering the cost of MATS was 3 See CAA section 112(n)(1)(A); see also Michigan flawed as it did not satisfy the EPA’s v. EPA, 135 S. Ct. at 2705 (‘‘Quite apart from the 4 U.S. EPA. 1998. Study of Hazardous Air obligation under CAA section hazardous-air-pollutants program, the Clean Air Act Pollutant Emissions from Electric Utility Steam 112(n)(1)(A), as interpreted by the U.S. Amendments of 1990 subjected power plants to Generating Units—Final Report to Congress, Supreme Court in Michigan. various regulatory requirements. The parties agree Volume 1. EPA–453/R–98–004a. that these requirements were expected to have the 5 In the same 2000 action, the EPA Administrator Additionally, the EPA proposed that collateral effect of reducing power plants’ emissions found that regulation of HAP emissions from while finalizing the action would of hazardous air pollutants, although the extent of natural gas-fired EGUs is not appropriate or reverse the 2016 Supplemental Finding, the reduction was unclear.’’). necessary. 65 FR 79826.

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00004 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations 31289

Both the 2005 Revision and the CAMR 2015, the U.S. Supreme Court held that In that analysis, the EPA estimated that were vacated by the D.C. Circuit in ‘‘EPA interpreted [CAA section the final MATS rule would yield total 2008. The Court held that the EPA had 112(n)(1)(A)] unreasonably when it annual monetized benefits (in 2007 failed to comply with the requirements deemed cost irrelevant to the decision to dollars) of between $37 billion to $90 of CAA section 112(c)(9) for delisting regulate power plants.’’ Michigan v. billion using a 3-percent discount rate source categories, and consequently also EPA, 135 S. Ct. 2699, 2712 (2015). In so and $33 billion to $81 billion using a 7- vacated the CAA section 111 holding, the U.S. Supreme Court found percent discount rate, plus additional performance standards promulgated in that the EPA ‘‘must consider cost— benefits that cannot be quantified, in CAMR, without addressing the merits of including, most importantly, cost of comparison to the projected $9.6 billion those standards. New Jersey v. EPA, 517 compliance—before deciding whether in annual compliance costs. That F.3d 574 (D.C. Cir. 2008). regulation is appropriate and analysis reflects that 99.9 percent of the In response to the New Jersey necessary.’’ Id. at 2711. It is ‘‘up the total annual monetized benefits were decision, the EPA conducted additional Agency,’’ the Court added, ‘‘to decide attributable not to benefits from HAP technical analyses, including peer- (as always, within the limits of reduction, but rather from benefits from reviewed risk assessments on human reasonable interpretation) how to co-reduction of non-HAP pollutants. In health effects associated with mercury account for cost.’’ Id. The rule was the 2016 Supplemental Finding, the and non-mercury HAP emissions from ultimately remanded back to the EPA EPA determined that both the preferred EGUs, focusing on risks to the most (without vacatur) to complete the ‘‘cost reasonableness’’ approach and the exposed and sensitive individuals in the required cost analysis. White Stallion alternative ‘‘cost benefit’’ approach population. Those analyses found that Energy Ctr. v. EPA, No. 12–1100, ECF supported the conclusion that mercury and non-mercury HAP No. 1588459 (D.C. Cir. December 15, regulation of HAP emissions from EGUs emissions from EGUs remain a 2015). is appropriate and necessary. significant public health hazard and that In response to the U.S. Supreme Several state and industry groups EGUs were the largest U.S. Court’s direction, the EPA in the 2016 petitioned for review of the 2016 anthropogenic source of mercury Supplemental Finding promulgated two Supplemental Finding in the D.C. emissions to the atmosphere.6 Based on different approaches to incorporate cost Circuit. Murray Energy Corp. v. EPA, these findings, in 2012, the EPA into the appropriate and necessary No. 16–1127 (D.C. Cir. filed April 25, affirmed the original 2000 Finding that finding. 81 FR 24420. The EPA’s 2016). In April 2017, the EPA moved the it is appropriate and necessary to preferred approach (referred to as the D.C. Circuit to continue oral argument regulate EGUs under CAA section 112. ‘‘cost reasonableness’’ approach) and hold the case in abeyance in order 77 FR 9304 (February 16, 2012). compared the estimated cost of to give the new Administration an In the same 2012 action, the EPA compliance in the regulatory impact opportunity to review the 2016 action. established a NESHAP, commonly analysis (RIA) for the 2012 MATS Final (As further explained below, as of the 7 called MATS, that required coal- and Rule (referred to here as 2011 RIA ) date of signature, the case remains oil-fired EGUs to meet HAP emission against several cost metrics relevant to pending in the D.C. Circuit.) standards reflecting the application of the EGU sector (e.g., historical annual Accordingly, the EPA reviewed the 2016 the maximum achievable control revenues, annual capital expenditures, action and proposed on February 7, technology (MACT) for mercury and and impacts on retail electricity prices). 2019, to correct flaws in the prior other air toxics. After MATS was The ‘‘cost reasonableness’’ approach did response to Michigan v. EPA (84 FR promulgated, both the rule itself and not compare costs to benefits. Under 2670). Specifically, the 2019 Proposal many aspects of the EPA’s appropriate this approach, the EPA concluded that proposed to reverse the 2016 action and and necessary finding were challenged the power sector would be able to to conclude that it is not ‘‘appropriate in the D.C. Circuit. In White Stallion comply with the MATS requirements and necessary’’ to regulate HAP Energy Center v. EPA, the Court denied while maintaining its ability to generate, emissions from coal- and oil-fired EGUs. all challenges. 748 F.3d 1322 (D.C. Cir. transmit, and distribute reliable The public comment period for the 2019 2014). One judge dissented, expressing electricity at reasonable cost to Proposal ended on April 17, 2019. The the view that the EPA erred by refusing consumers. Using a totality-of-the- remainder of this section of this to consider cost in its ‘‘appropriate and circumstances approach, the EPA preamble responds to significant necessary’’ determination. Id. at 1258– weighed this analysis that the costs of comments received on the appropriate the rule were reasonable along with its 59 (Kavanaugh, J., dissenting). and necessary finding and describes the The U.S. Supreme Court subsequently prior findings about the amount of HAP EPA’s justification for finalizing this granted certiorari, directing the parties pollution coming from the Coal- and reversal of the 2016 Supplemental to address a single question posed by Oil-Fired EGU source category, the Finding. the Court itself: ‘‘Whether the scientific studies and modeling assessing the risks to public health and C. EPA’s Finding Under CAA Section Environmental Protection Agency the environment from domestic EGU 112(n)(1)(A) unreasonably refused to consider cost in HAP pollution, and information about determining whether it is appropriate to 1. EPA Has the Statutory Authority To the toxicity and persistence of HAP in regulate hazardous air pollutants Revisit the Appropriate and Necessary the environment. Finding emitted by electric utilities.’’ Michigan In a second, alternative, and v. EPA, 135 S. Ct. 702 (Mem.) (2014). In independent approach (referred to as a. Summary of 2019 Proposal the ‘‘cost benefit’’ approach), the EPA Section 112(n)(1)(A) of the CAA 6 U.S. EPA. 2011. Revised Technical Support Document: National-Scale Assessment of Mercury considered the benefit-cost analysis in directs the Administrator of the EPA to Risk to Populations with High Consumption of Self- the RIA for the 2012 MATS Final Rule. determine whether it is ‘‘appropriate caught Freshwater Fish in Support of the and necessary’’ to regulate HAP Appropriate and Necessary Finding for Coal- and 7 U.S. EPA. 2011. Regulatory Impact Analysis for emissions from fossil fuel-fired EGUs Oil-Fired Electric Generating Units. Office of Air the Final Mercury and Air Toxics Standards. EPA– after conducting a study of the hazards Quality Planning and Standards. December. EPA– 452/R–11–011. Available at: https://www3.epa.gov/ 452/R–11–009. Docket ID Item No. EPA–HQ–OAR– ttn/ecas/docs/ria/utilities_ria_final-mats_2011- to public health reasonably anticipated 2009–0234–19913. 12.pdf. to occur as a result of emissions of HAP

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00005 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 31290 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations

from EGUs after imposition of emission finding that has no regulatory effect rethinking its interpretation of such controls imposed under other would be ‘‘inherently irrational,’’ and provision. The EPA’s CAA rulemaking provisions of the CAA. In Michigan v. that the EPA has failed to articulate a history contains many examples of the EPA, the U.S. Supreme Court instructed reasoned basis for undertaking this Agency’s changing position on a the Agency that it was required to action (citing Air Alliance Houston v. previous interpretation of a provision, consider cost as part of its appropriate EPA, 906 F.3d 1049 (D.C. Cir. 2018), and even where there is no explicit directive and necessary determination. The asserting that in that decision the D.C. within the provision to review or revise. Agency completed a consideration of Circuit found an EPA rule irrational Absent a specific statutory the cost to regulate HAP emissions from where the EPA tried to ‘‘have it both prohibition, the EPA’s ability to revisit coal- and oil-fired EGUs in the 2016 ways’’ by claiming that a rule was existing decisions is well established. Supplemental Finding. The EPA’s 2019 necessary to prevent harms to regulated The EPA has inherent authority to action proposed to revisit the 2016 industry but also ‘‘does nothing more reconsider and/or revise past decisions Supplemental Finding’s consideration than maintain the status quo,’’ Id. at to the extent permitted by law so long of cost, on the basis that the 2016 action 1068). as the Agency provides a reasoned is flawed. The 2019 Proposal stated that Other commenters said that the EPA explanation. The authority to reconsider such reexamination was permissible as has authority to reconsider prior Agency exists in part because the EPA’s a basic principle of administrative law decisions and the 2016 Supplemental interpretations of statutes it administers and under the CAA. 84 FR 2674 n.3. Finding in particular. These ‘‘[are not] instantly carved in stone,’’ but commenters noted that if the 2016 must be evaluated ‘‘on a continuing b. Final Rule Supplemental Finding were left basis.’’ Chevron U.S.A. v. Natural The EPA is finalizing this action as unamended, it would establish policy Resources Defense Council, 467 U.S. proposed in February 2019 on the basis precedents at odds with well- 837, 863–64. This is true when, as is the that the CAA and CAA section established precepts about how benefits case here, review is undertaken partly 112(n)(1)(A) do not prohibit the and costs should be considered in ‘‘in response to . . . a change in Administrator from revisiting a prior regulatory decisions. administrations.’’ National Cable & finding made under that section. Response: The EPA disagrees with Telecommunications Ass’n v. Brand X commenters that CAA section c. Comments and Responses internet Services, 545 U.S. 967, 981 112(n)(1)(A) speaks to the EPA’s (2005). Indeed, ‘‘[a]gencies obviously Comment: Some commenters asserted authority to revisit its appropriate and have broad discretion to reconsider a that it is unlawful for the EPA to revisit necessary finding, and we, therefore, regulation at any time.’’ Clean Air its 2016 Supplemental Finding at all, disagree with commenters’ contention Council v. Pruitt, 862 F.3d 1, 8–9 (D.C. because the EPA has completed the that the statute on its face prohibits the Cir. 2017). analytic process Congress set in motion EPA from revisiting a determination Commenters’ assertions that the in 1990, and the statute unambiguously made under that provision. The statutory context and structure of CAA prohibits the EPA from revisiting or provision reads: ‘‘The Administrator section 112 and the legislative history of revising the CAA section 112(n)(1)(A) shall regulate electric utility steam that provision support their view that finding. Commenters asserted that the generating units under this section, if the EPA lacks authority to revisit its legislative history, statutory context, the Administrator finds such regulation CAA section 112(n)(1)(A) determination and statutory structure support their is appropriate and necessary after are marred by the commenters’ assumed position that Congress intended the considering the results of the study premise that the EPA necessarily would CAA section 112(n)(1)(A) appropriate required by this subparagraph [the find that it is appropriate and necessary and necessary finding to be a one-time ‘‘Utility Study’’ 8].’’ The only clear to regulate EGUs. The commenters argue decision, and that the provision gives requirement with regard to timing or that their interpretation of the statute the EPA ‘‘limited discretion to activate sequence found in the text of the must be correct because it creates a tidy a one-way switch to ‘turn on’ regulation provision is that the Administrator may framework: The EPA makes an of power plants.’’ The commenters not make the finding prior to affirmative appropriate and necessary argued that ‘‘[o]nce EPA turns on that considering the results of the Utility finding, regulations under CAA section switch, as it did in its 2000 finding . . . Study, which the EPA completed in 112 are promulgated, and the only it must regulate power plants under 1998. The statute does not restrict the statutory means by which the section 112.’’ Administrator’s ability to revise or appropriate and necessary finding could Moreover, those commenters argued reconsider a prior finding made under be revisited is to satisfy the delisting that even if CAA section 112 were CAA section 112(n)(1)(A). criteria under CAA section 112(c)(9). ambiguous as to the EPA’s authority to We also disagree with commenters’ According to commenters, such a revisit the appropriate and necessary argument that because other statutory framework fits with Congress’ concerns finding, the EPA was still bound to provisions in the CAA mandate that the about dangers to public health and follow CAA section 112(c)(9)’s delisting EPA review and revise regulations on a welfare due to air pollution and what procedure before it could reverse its set schedule or continuing basis, it must they broadly characterize as finding under CAA section 112(n)(1)(A). follow that every other statutory congressional desire to regulate HAP The commenters claimed that New provision lacking such a review-and- from power plants ‘‘promptly.’’ The Jersey confirms that the EPA lacks revise clause prohibits an agency from problem with the commenters’ statutory inherent authority to reconsider the interpretation is that it makes sense only appropriate and necessary finding. 8 CAA section 112(n)(1)(A) directs the EPA to if an affirmative appropriate and Finally, the commenters claimed that conduct a study to evaluate the hazards to public necessary finding occurs in the first health reasonably anticipated to occur as the result it would be ‘‘illogical’’ for the EPA to of HAP emissions from EGUs after the imposition instance. If, as commenters assert, CAA have authority to revise the appropriate of the requirements of the CAA, and to report the section 112(c)(9) is the only statutory and necessary finding independent of results of such study to Congress by November 15, means by which a finding under CAA removing power plants from the list of 1993. See U.S. EPA, Study of Hazardous Air section 112(n)(1)(A) may be revisited, Pollutant Emissions from Electric Utility Steam regulated sources under CAA section Generating Units—Final Report to Congress. EPA– commenters’ framework provides no 112. Commenters argued that a revised 453/R–98–004a, February 1998. pathway by which the EPA could revisit

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00006 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations 31291

a finding that it is not appropriate and into CAA section 112(n)(1)(A)’s criticizing the EPA for attempting to necessary to regulate HAP from power appropriate and necessary characterize its rule as relieving plants. Commenters’ ‘‘unambiguous’’ determination, such that a failure to ‘‘substantial compliance and reading of CAA section 112(n)(1)(A) and consider those criteria in the context of implementation burden’’ while also its assumption that Congress drafted the reversing a determination under CAA ‘‘maintaining the status quo’’ (such that provision in order to ensure ‘‘prompt’’ section 112(n)(1)(A) is arbitrary and the rule would have little effect on reductions of HAP from EGUs treats an renders CAA section 112(c)(9) a nullity. compliance requirements). See Air affirmative finding under that section as As explained in section II.D of this Alliance Houston, 906 F.3d at 1068. a foregone conclusion rather than a preamble, we agree that the EPA may Here, the Agency believes a different decision left up to the expertise of the not delist EGUs from the CAA section finding and better response to the U.S. Agency and its Administrator. 112(c) list and revoke MACT standards Supreme Court’s decision in Michigan The commenters’ reading of the for power plants without meeting the v. EPA is warranted given the proper statute also cannot be squared with the delisting criteria of CAA section application of that decision and the Michigan v. EPA decision. They assert 112(c)(9). We do not agree, however, facts in the EPA’s record. We that CAA section 112(n)(1)(A) only that the delisting provision has any acknowledge that this change in policy allows the EPA ‘‘to activate a one-way effect on the Agency’s ability to make an will not affect the CAA section 112 switch to ‘turn on’ regulation,’’ and affirmative or negative determination MACT standards for EGUs because the notes that the Agency did so ‘‘in its under CAA section 112(n)(1)(A) where D.C. Circuit’s decision in New Jersey v. 2000 finding.’’ Commenters are we are not purporting to alter the CAA EPA prohibits the Agency from essentially arguing that the U.S. section 112(c) list. In particular, we do removing listed sources from the CAA Supreme Court’s instruction to the EPA not agree with the commenters’ reading section 112(c) list without satisfying the that it was required to consider cost as of New Jersey that the D.C. Circuit’s CAA section 112(c)(9) delisting criteria part of a CAA section 112(n)(1)(A) holding means that the EPA could (see section II.D of this preamble). But finding could never have had any reverse an affirmative appropriate and we do not agree that simply because practical effect, because according to necessary finding only if it found that D.C. Circuit precedent establishes that commenters, the ‘‘only . . . statutorily the CAA section 112(c)(9) delisting the Agency’s reversing its prior mandated avenue to turn the switch off criteria were met. The Court’s holding determination will have a particular and reverse course . . . [is] the section in New Jersey plainly states that CAA regulatory consequence, the Agency is, 112(c)(9) procedures.’’ Therefore, in section 112(c)(9) ‘‘unambiguously therefore, prohibited from revisiting that petitioners’ view, regardless of what the limit[s] EPA’s discretion to remove prior determination in the first instance. EPA determined on remand from sources, including EGUs, from the Comment: Some commenters stated Michigan, only the satisfaction of the section 112(c)(1) list once they have that the EPA has no authority to CAA section 112(c)(9) criteria, which been added to it.’’ 517 F.3d 574, 583 ‘‘revise’’ its response to the U.S. contain no consideration of cost, could (D.C. Cir. 2008). Commenters’ presumed Supreme Court’s decision in Michigan, have altered the EPA’s finding under incorporation of the statutory delisting and its attempt to do so would CAA section 112(n)(1)(A). We do not criteria into the CAA section impermissibly subvert the judicial agree that this is a reasonable reading of 112(n)(1)(A) determination also finds no review process. These commenters the statute or the Michigan decision. support in the Michigan decision, which argued that the EPA’s response to Additionally, the EPA notes that the said nothing about the EPA’s obligation Michigan is the 2016 Supplemental D.C. Circuit in New Jersey held that the to consider those criteria in determining Finding, and that at this stage, that EPA’s reversal of a prior determination whether regulation of power plants is response cannot be altered or reversed. that it was appropriate and necessary to appropriate and necessary. The commenters contended that the regulate EGUs under CAA section 112 Finally, we disagree with commenters 2016 Supplemental Finding constitutes did not by itself effect a delisting of who assert that this final action is final Agency action and noted that the EGUs from the CAA section 112(c) list ‘‘inherently irrational’’ because the Finding is currently subject to petitions of source categories. This holding MATS standards would not be reversed for review in the D.C. Circuit. The recognizes that the CAA section 112 as a result of the negative appropriate commenters suggested that seeking to appropriate and necessary and necessary finding, due to undo the 2016 Supplemental Finding by determination is structurally and controlling legal precedent from the administrative action would unlawfully functionally separate from the EPA’s D.C. Circuit (New Jersey). In this action circumvent that review. Other ability, conditioned on certain predicate the EPA is setting out the Agency’s commenters asserted that the EPA has findings, to remove source categories revised reasoning to respond to a U.S. an obligation to explain how final action from the CAA section 112(c) list. Supreme Court decision and remand on the 2019 Proposal could impact the Commenters are, therefore, wrong to (Michigan), because the EPA concludes government’s position in ongoing assert that the EPA can reverse an that the 2016 Supplemental Finding is litigation of the 2016 Supplemental appropriate and necessary not appropriate as a matter of Finding. Commenters also said the EPA determination under CAA section interpretation of the statute or as a must address the implications of a 112(n)(1)(A) only if it has first matter of policy. As noted by some of reversal of that finding, considering the undertaken CAA section 112(c)(9)’s the commenters, leaving in place the petitioner’s positions in the ongoing delisting procedure, and wrong to assert incorrect interpretation of ‘‘appropriate’’ litigation where the petitioner has that New Jersey supports their position in CAA section 112(n)(1)(A) could argued that reversal of the appropriate that the EPA lacks inherent authority to establish policy precedent that could and necessary finding must be followed reconsider the appropriate and have ‘‘long-term and harmful by vacatur of MATS. necessary finding; in fact, that case consequences.’’ Response: The EPA disagrees with the supports the opposite position. Moreover, the EPA disagrees that Air commenters that finalizing this action For similar reasons, we also reject the Alliance Houston v. EPA has any ‘‘subverts the judicial review process’’ commenters’ contention that CAA bearing on this action. There, in with respect to the 2016 Supplemental section 112(c)(9)’s health protective admonishing the Agency that it could Finding. To the extent that commenters criteria are substantively incorporated not ‘‘have it both ways,’’ the Court was are arguing that the EPA lacks statutory

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00007 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 31292 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations

authority to review the 2016 such litigation positions in this final Treating power plants differently from Supplemental Finding, the EPA has action. other sources.’’) (emphasis in original). addressed that contention in the c. Comments and Responses response to the comment above. We 2. The Preferred Cost Reasonableness agree that the 2016 Supplemental Approach of the 2016 Supplemental Comment: Some commenters asserted Finding constituted final Agency action, Finding Was Deficient that the cost analysis in the 2016 and we acknowledge that petitions for a. Summary of 2019 Proposal Supplemental Finding was consistent review of that action were filed in the with longstanding cost-effectiveness D.C. Circuit in Murray Energy Corp. v. The EPA proposed to determine that methodologies used in other CAA EPA, No. 16–1127 (and consolidated the Agency’s 2016 Supplemental programs, such as the CAA section 111 cases) (D.C. Cir. filed April 25, 2016). Finding erred in its consideration of New Source Performance Standards and However, we disagree that our final cost. Specifically, we proposed to find CAA section 169 Prevention of action unlawfully circumvents the that what was described in the 2016 Significant Deterioration (PSD). These judicial process. The EPA filed a motion Supplemental Finding as the preferred commenters disagreed with what they in the Murray Energy litigation approach, or the ‘‘cost reasonableness characterized as the 2019 Proposal’s requesting the Court to continue oral test,’’ does not meet the statute’s position that CAA section 111 case law argument, which had been scheduled requirements to fully consider costs and was irrelevant to the CAA section for May 18, 2017, to allow the new was an unreasonable interpretation of 112(n)(1)(A) appropriate and necessary Administration adequate time to review CAA section 112(n)(1)(A)’s mandate, as determination, noting that cost the 2016 Supplemental Finding to informed by the U.S. Supreme Court’s effectiveness is used in CAA section 111 determine whether it needed to be opinion in Michigan. A summary of that to determine standards for existing reconsidered.9 On April 27, 2017, in approach can be found in the 2019 sources, much as the EPA is consideration of the EPA’s motion, the Proposal. 84 FR 2674–75. determining whether to regulate existing D.C. Circuit ordered that the sources in CAA section 112(n)(1)(A). b. Final Rule consolidated challenges to the 2016 These commenters further said that the Supplemental Finding be held in After considering comments proposed monetized cost-benefit abeyance.10 That case continues to be submitted in response to the EPA’s 2019 approach is inferior to the longstanding held in abeyance, pending further order Proposal, the EPA is finalizing the cost-effectiveness test for addressing of the Court. In its order, the Court concerns about standards that impose directed the parties to file motions to proposed approach. The EPA concludes that the ‘‘preferred approach’’ in the costs too high for the industry to bear. govern future proceedings within 30 However, other commenters agreed with days of the Agency’s concluding its 2016 Supplemental Finding did not meaningfully consider cost, which the the EPA that cases interpreting section review of the 2016 Supplemental 111 of the CAA were not an appropriate 11 Michigan Court observed to be a Finding. guide to considering costs under CAA The EPA disagrees with the ‘‘centrally relevant factor’’ in making the section 112(n)(1)(A). commenters that the Agency has an CAA section 112(n)(1)(A) appropriate obligation to address in the context of and necessary finding. The 2016 Response: The broad language of CAA this regulatory action the government’s Supplemental Finding’s de-emphasis of section 112(n)(1)(A) and the holding of position in that ongoing litigation. We the importance of the cost consideration the Michigan Court suggest that there is address in section II.D of this preamble in the appropriate and necessary more than one permissible way to the implications of the reversal of the determination was based on an interpret the Agency’s obligation to 2016 Supplemental Finding, including impermissible attempt to ‘‘harmonize’’ consider cost in the appropriate and addressing those comments received CAA section 112(n)(1)(A) with the necessary finding. The text of that that argue that a vacatur of MATS is remainder of CAA section 112,12 and section does not require the Agency to required upon finalization of this action. was not consistent with Congress’ intent consider cost in a particular fashion. To the extent that the commenter is and the U.S. Supreme Court’s decision The U.S. Supreme Court, in identifying suggesting that it would be appropriate in Michigan v. EPA, given that statutory that the Agency’s obligation to consider or required for the EPA at this point to provision’s directive to treat EGUs cost in some fashion in light of the address potential future arguments differently from other sources. See 135 broad term ‘‘appropriate,’’ recognized petitioners might make in the Murray S. Ct. at 2710 (‘‘The Agency claims that the discretion afforded the Energy litigation following this final it is reasonable to interpret [CAA Administrator, noting, ‘‘[i]t will be up action, the Agency disagrees. The section 112(n)(1)(A)] in a way that the Agency to decide (as always, within appropriate venue for addressing such ‘harmonizes’ the program’s treatment of the limits of reasonable interpretation) arguments is the judicial review process power plants with its treatment of other how to account for cost.’’ 135 S. Ct. at for that action. Commenters provide no sources. This line of reasoning 2711. Even in the final 2016 authority to support their assertion that overlooks the whole point of having a Supplemental Finding, the EPA an agency is obliged to discuss in a separate provision about power plants: acknowledged that the cost rulemaking the implications of that reasonableness test was but one way to rulemaking for pending litigation 12 See Legal Memorandum Accompanying the interpret its CAA section 112(n)(1)(A) challenging a previous, related agency Proposed Supplemental Finding that it is obligation to consider cost, and ‘‘that action; the EPA is aware of no such Appropriate and Necessary to Regulate Hazardous the agency need not demonstrate that authority; and the EPA declines to take Air Pollutants from Coal- and Oil-Fired Electric [its] decision is the same decision that Utility Steam Generating Units (EGUs) (2015 Legal Memorandum) (Docket ID Item No. EPA–HQ–OAR– would be made by another 9 Respondent EPA’s Motion to Continue Oral 2009–0234–20519), at 6–15 (describing statutory Administrator or a reviewing court.’’ 81 Argument at 6, Murray Energy Corp. v. EPA, No. purpose of 1990 CAA Amendments and CAA FR 24431. The commenters provide 16–1127 (D.C. Cir. April 18, 2017), ECF No. section 112, and concluding that ‘‘while cost is many reasons for why they preferred the 1671687. certainly an important factor, it is one of several 10 Order, Murray Energy Corp. v. EPA, No. 16– factors that must be considered and section EPA’s ‘‘cost reasonableness’’ test, but 1127 (D.C. Cir. April 27, 2017), ECF No. 1672987. 112(n)(1) does not support a conclusion that cost even they do not attempt to argue that 11 Id. should be the predominant or overriding factor.’’). the EPA’s 2016 ‘‘preferred approach’’ is

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00008 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations 31293

the only permissible interpretation of by the courts.’’ Id. at 218 (emphasis in the harmful effects of HAP, and, the statute. original). therefore, shares the section’s overall Comparisons of a regulation’s costs The EPA’s choice to employ cost- focus on harm prevention. These effectiveness analyses, rather than cost- and the relationship of those costs to the commenters asserted that the ‘‘preferred benefit comparisons, in the context of benefits the regulation is expected to approach’’ in the 2016 Supplemental other statutory provisions such as CAA Finding met the requirements of accrue are a traditional and section 111 or the PSD program in no Michigan and were consistent with commonplace way to assess the costs of way binds the Agency to using that congressional intent and the CAA’s a regulation and are a permissible way method to consider cost in CAA section statutory goals. to comply with Congress’ broad 112(n)(1)(A). The EPA’s citation in the Other commenters, however, agreed directive to the Administrator to 2015 Legal Memorandum of our with the 2019 Proposal that the ‘‘cost determine whether regulation is consideration of cost under CAA section reasonableness’’ test in the 2016 ‘‘appropriate’’ in CAA section 111 and the case law evaluating those Supplemental Finding’s ‘‘preferred 112(n)(1)(A). The EPA has never taken instances was only to provide context to approach’’ was invalid, harmful, and the position, nor do commenters argue explain the genesis of the EPA’s newly failed to meet the Michigan Court’s now, that any comparison of costs to minted ‘‘cost reasonableness’’ test in the expectation that the Agency should benefits would be an impermissible 2016 Supplemental Finding. Even then weigh benefits against costs. These reading of the Agency’s obligation to the EPA did not take the position that commenters characterized the cost- consider cost in CAA section the D.C. Circuit cases reviewing the reasonableness test, which compared 112(n)(1)(A); indeed, the Agency’s Agency’s cost considerations under costs of MATS compliance with various alternative approach to considering cost CAA section 111 were binding other costs incurred by the power in the 2016 Supplemental Finding was precedent upon which the Court should sector, as an ‘‘affordability test,’’ or an a formal cost-benefit analysis based on review our action under CAA section inquiry into whether the power sector its 2011 RIA, and many of the 112(n)(1)(A). In short, the commenters’ could absorb the costs of compliance. commenters who now evince a preference that the EPA consider cost in These commenters noted that such a test preference for the 2016 ‘‘cost a different way does not preclude the ignores benefits by failing to provide reasonableness test’’ at the time agreed Agency from instead considering cost important information on whether that the 2011 RIA cost-benefit analysis using an approach that compares costs society’s investment in additional costs could independently satisfy the and benefits, where the statute’s broad is worth the expected benefits and fails Agency’s obligation to consider cost directive suggests that it may. See to consider whether costs would be under CAA section 112(n)(1)(A). U.S. Entergy, 556 U.S. at 226. ‘‘prudently incurred’’ as a means to Supreme Court precedent also supports Comment: Some commenters asserted reduce hazards to public health. As one the Agency’s position that, absent an that the EPA’s proposed approach to commenter put it, ‘‘Simply because the unambiguous prohibition to use cost- considering costs and benefits is power sector could absorb costs without benefit analysis, the Agency generally inconsistent with what they broadly affecting current operational characterize as congressional intent to may do so as a reasonable way to performance does not mean that it err on the side of protecting public consider cost.13 In Entergy Corp. v. should absorb those costs.’’ Some health. These commenters argued that commenters objecting to the ‘‘preferred Riverkeeper, Inc., 556 U.S. 208 (2009), Congress recognized the insufficiency of approach’’ in the 2016 Supplemental the U.S. Supreme Court struck down a available methods for quantifying costs Finding emphasized that looking at cost Second Circuit decision prohibiting the and benefits when revising CAA section in this manner would invite the EPA from employing benefit-cost 112 in 1990 and that Congress promulgation of regulations that are analysis where the statute was silent as concluded that the nature and latency of poorly designed, with few potential to how the Agency was to consider cost harms posed by HAP are not given benefits. They voiced concern that using in adopting standards for cooling water sufficient weight in a regulatory process affordability tests could result in intake standards for power plants. The that must balance long-term benefits agencies focusing public and private Second Circuit found that because against present-day costs. Commenters sector resources on extinguishing analogous provisions in the Clean Water said that the Agency should not relatively small risks while leaving Act explicitly instructed the EPA to construe the Michigan Court’s larger risks unattended. Other consider ‘‘the total cost of application of instruction to ‘‘meaningfully consider commenters noted that such tests also technology in relation to the effluent cost’’ as a requirement to consider penalize successful industries due to reduction benefits to be achieved,’’ (33 benefits in a way that is inconsistent their success, and risk failing to U.S.C. 1314(b)(4)(B)), Congress’ failure with Congress’ determination that appropriately regulate industries that to include such an instruction to the reductions in HAP emissions have great are less profitable. EPA in the provision at issue in the case value to the public. These commenters Response: The EPA agrees with meant that the EPA was not permitted added that the EPA’s proposed commenters who stated that Congress’ to compare compliance costs to approach is based on an incorrect intent with respect to CAA section 112, expected environmental benefits. The interpretation of Michigan, which stated as a whole, evinces an acknowledgment U.S. Supreme Court reversed, holding only that consideration of cost should of the seriousness of toxic air pollutants. that the EPA’s use of cost-benefit play some role in the appropriate and We do not agree, however, that general analysis ‘‘governs if it is a reasonable necessary finding, not that cost congressional concern about the toxicity interpretation of the statute—not considerations should dominate that of HAP overrides the specific necessarily the only possible finding. According to these commenters, instruction given to the Administrator interpretation, nor even the the studies required in CAA section in CAA section 112(n)(1)(A) to make a interpretation deemed most reasonable 112(n) indicate that Congress put public determination about whether regulation health and environmental concerns at of EGUs in particular is ‘‘appropriate 13 See S. Masur & Eric A. Posner, Cost-Benefit the forefront of CAA section 112, which and necessary.’’ As the U.S. Supreme Analysis and the Judicial Role, 85 U. Chi. L. Rev. was enacted explicitly in response to Court admonished the EPA in Michigan, 935, 981 (2018). the EPA’s lack of action in addressing the text and structure of CAA section

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00009 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 31294 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations

112, and 112(n)(1)(A) in particular, on a formal benefit-cost analysis, but prevented. See 81 FR at 24432. In other evince Congressional design to neither did it hold that a comparison of words, much as it did in 2012 when it approach the question whether to costs and benefits is an impermissible read cost consideration entirely out of regulate EGUs differently than other approach to considering cost. the CAA section 112(n)(1)(A) source categories: The U.S. Supreme Court determination, the Agency in 2016 was Congress crafted narrow standards for EPA contemplated that a proper fixated on the term ‘‘necessary,’’ to apply when deciding whether to regulate consideration of cost would be relative without considering whether any other sources; in general, these standards to benefits, and the Court’s decision countervailing factors, i.e., cost, might concern the volume of pollution emitted by contains many references comparing the call into question whether regulation the source, [CAA section 112(c)(1)], and the two considerations. In establishing the was ‘‘appropriate.’’ As many threat posed by the source ‘‘to human health facts of the case, the Court pointed out commenters pointed out, the ‘‘cost or the environment,’’ [citing CAA section that ‘‘EPA refused to consider whether reasonableness test’’ failed to consider 112(c)(3)]. But Congress wrote the provision the costs of its decision outweighed the cost relative to benefits, and really before us [CAA section 112(n)(1)(A)] more benefits.’’ 135 S. Ct. at 2706. The Court focused only on whether costs could be expansively . . . That congressional election questioned whether a regulation could settles this case. [The Agency’s] preference absorbed, rather than on whether they for symmetry cannot trump an asymmetrical be considered ‘‘rational’’ where there should be absorbed—the inquiry that is statute. was a gross imbalance between costs specifically required by the word and benefits and stated that ‘‘[n]o ‘‘appropriate.’’ We, therefore, conclude 135 S. Ct. at 2710 (internal citations regulation is ‘appropriate’ if it does that the ‘‘cost reasonableness’’ approach omitted). more harm than good.’’ Id. at 2707. The did not adequately address the U.S. Moreover, we do not agree with Court also made numerous references to Supreme Court’s instruction that a commenters’ suggestion that in the a direct comparison of the costs of reasonable regulation requires an agency Agency’s comparison of costs and MATS with benefits from reducing to fully consider ‘‘the advantages and benefits, the EPA is considering benefits emissions of HAP. For instance, the the disadvantages’’ of a decision. See in a way that is inconsistent with a Court pointed out that ‘‘[t]he costs [of Michigan, 135 S. Ct. at 2707 (emphasis congressional determination that MATS] to power plants were thus in original). reductions in HAP emissions have great between 1,600 and 2,400 times as great Moreover, we take seriously value to the public and Congress’ public as the quantifiable benefits from commenters’ concerns that leaving the health and environmental concerns. We reduced emissions of hazardous air ‘‘preferred approach’’ in place, with its disagree that CAA section 112’s general pollutants.’’ Id. at 2706. Although the ‘‘cost reasonableness’’ or affordability concerns about public health and Court’s holding established no bright- test, could have a harmful influence on environmental risks from HAP line rules, the opinion as a whole, thus, other agencies interpreting similarly emissions mandated a particular repeatedly suggests that CAA section broad congressional directives to manner of valuing or weighing the 112(n)(1)(A)’s requisite consideration of consider cost. Statutes that direct benefits of reducing those risks. cost would not be met if the cost agencies to make determinations about As noted in the 2019 Proposal, we do analysis did not ‘‘prevent the imposition whether regulation is ‘‘appropriate’’ are not think the 2016 Supplemental of costs far in excess of benefits.’’ Id. at precisely the contexts in which those Finding’s analysis of cost satisfied the 2710. agencies should retain discretion to Agency’s mandate under CAA section The 2016 Supplemental Finding’s select and prioritize public policies 112(n)(1)(A) and Michigan. The ‘‘test’’ of whether an industry can bear which provide the most value for the ‘‘preferred approach’’ in the 2016 the cost of regulation, and its public good in relation to the cost. Supplemental Finding considered cost subsequent conclusion that such costs Comment: Commenters said that the insofar as the Agency at the time are ‘‘reasonable,’’ does not satisfy the EPA’s proposed new approach to analyzed whether the utility industry as statute’s mandate to determine whether considering cost in the CAA section a whole could continue to operate, and such regulation is appropriate and 112(n)(1)(A) finding is an impermissible found that it could (i.e., that costs were necessary. We agree with commenters interpretation of that provision because ‘‘reasonable’’). 81 FR 24420, 24422, who stated that the metrics ‘‘tested’’ by it fails to meaningfully address factors 24424, 24427, 24428, 24429, 24430, the Agency in the 2016 Supplemental that are ‘‘centrally relevant’’ to the 24431. But we do not think the Finding are not an appropriate basis for inquiry of whether it is appropriate and ‘‘preferred approach’’ in the 2016 the determination whether it is necessary to regulate HAP from EGUs. Finding gave sufficient weight to cost as ‘‘appropriate and necessary’’ to impose Some commenters noted that the a ‘‘centrally relevant factor,’’ Michigan, that regulation. Each cost metric the Agency’s alleged failure in the 2019 135 S. Ct. at 2707—that is, we do not Agency examined compared the cost of Proposal to adequately address these think that a cost standard that is MATS to other costs borne by the factors, upon which the 2016 satisfied by establishing that regulation industry, but never in its ‘‘preferred Supplemental Finding was predicated, will not fundamentally impair the approach’’ did the Agency make the runs afoul of the Agency’s obligation to functioning of a major sector of the assessment of whether the benefits provide a reasoned explanation for economy places cost at the center of a garnered by the rule were worth it—i.e., abandoning these considerations, citing regulatory decision—and we are in this a comparison of costs and benefits. Even Motor Vehicle Mfrs. Ass’n of United action heeding the Michigan Court’s if the EPA determined that cost of States, Inc. v. State Farm Mut. reading of the Administrator’s role regulation was, viewed on its own Automobile Ins. Co., 463 U.S. 29 (1983) under CAA section 112(n)(1)(A), which terms, unreasonable after comparing the and FCC v. Fox Television Stations, Inc., directed the Agency to meaningfully cost of regulation to other costs borne by 556 U.S. 502 (2009). The commenters consider cost within the context of a the industry, the ‘‘preferred approach’’ noted that these cases state the principle regulation’s benefits. We agree that could have still resulted in a finding that agencies cannot simply ignore prior Michigan did not hold that the Agency that regulation was ‘‘appropriate’’ factual determinations but must provide is required to base its decision whether because the EPA placed so much weight a ‘‘reasoned explanation’’ for a proposed it is appropriate and necessary to on hazards to public health and the departure from ‘‘facts and circumstances regulate EGUs under CAA section 112 environment that needed to be that underlay or were engendered by the

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00010 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations 31295

prior policy.’’ These commenters tribal members, leading to permanent CAA section 112(n)(1)(A), authorized specifically faulted the EPA for not cultural loss. Furthermore, these the EPA to reassess the benefits of giving appropriate weight to the commenters stated that many tribes are reducing those harms in the context of following factors: connected to particular waters for EGUs. The commenters stated that no cultural, spiritual, or other reasons (and study, including the EPA’s Utility i. Unquantified Benefits others’ fishing rights are limited to Study, suggests that HAP from EGUs are Commenters stated that the 2019 certain grounds by treaty), so tribal of any different character or pose less Proposal does not acknowledge that members cannot simply move their harm by their nature than HAP emitted some ‘‘hazards to public health’’ are fishing to another location to avoid by any other industrial source category. unquantified and asserted that the 2019 mercury contamination. The iv. Distributional Impacts of the Proposal presents a significant change commenters asserted that the preferred Pollutants on the Population in position with insufficient approach of the 2016 Supplemental justification for revising the EPA’s Finding recognized that regulation of Commenters pointed to Congress’ longstanding interpretation that the HAP from EGUs would benefit intent to address harms that are phrase ‘‘hazards to public health’’ American Indians by allowing them to concentrated within particular encompasses risks that have not been safely engage in, and thereby communities or populations, citing CAA monetized because of the limitations of perpetuate, their culture. These section 112(f)(2)(A)’s requirement that current methods, data, and uncertainty. commenters argued that the Agency’s the EPA address lifetime excess cancer Commenters said the 2019 Proposal preferred approach in the 2016 risks borne by the ‘‘individual most gave no discernable weight to these Supplemental Finding properly deemed exposed to emissions,’’ CAA section risks as required by the statutory phrase these qualitative benefits to be 112(n)(1)(C)’s directive that the EPA ‘‘hazards to public health reasonably cognizable and highly significant. In consider power plant mercury harms to anticipated to occur.’’ addition, the commenters stated that sensitive fish-consuming populations, Moreover, the commenters asserted mercury emissions likewise cause and legislative history (‘‘EPA is to that the monetized, HAP-specific significant harm to Indian subsistence consider individuals who are sensitive benefits at issue, which quantify and fishing economies, contaminating to a particular chemical’’ in assessing avoided IQ loss in children associated food sources that many tribal members whether a pollutant’s harm warrants with prenatal methylmercury exposure depend on for survival. According to regulation) (Leg. Hist. at 8501). The from self-caught fish consumption these commenters, the EPA’s 2016 commenters noted that the 2016 among recreational anglers, are but a preferred approach methodology Supplemental Finding’s preferred small fraction of the public health allowed for a full range of qualitative approach identified several populations benefits attributable to reductions in benefits to be accounted for, whereas that were disproportionately at risk of mercury emissions alone. The the 2019 proposed reversal does not. mercury exposure from EGUs, including commenters cited the statement from African-Americans living below the the EPA’s Science Advisory Board iii. Latency, Persistence in the poverty line in the Southeast who rely (SAB), which stated that IQ loss ‘‘is not Environment, and Toxicity of Regulated on the fish they catch for food, and the the most potentially significant health Pollutants children and fetuses in those effect associated with mercury exposure Some commenters asserted that the communities in particular whose risk of as other neurobehavioral effects, such as EPA’s proposed approach disregarded exposure is amplified; and individuals language, memory, attention, and other the physiochemical nature and toxicity and communities who live near coal- developmental indices, are more of the toxic air pollutants regulated by and oil-fired power plants, who are responsive to mercury exposure.’’ 80 FR CAA section 112 and the concern disproportionately members of racial 75040. The commenters noted that none Congress had expressed about these and ethnic minorities. The commenters of the environmental benefits from qualities in enacting that section. These cited a study that found that of the 8.1 reductions in mercury emissions could commenters pointed out that, in million people living within 3 miles of be quantified, nor any of the health or enacting the list of regulated air toxics, a coal-fired plant in the year 2000, 39 environmental benefits attributable to Congress deliberately withdrew the percent were people of color, a reductions in other HAP. EPA’s authority to judge the importance percentage significantly higher than the of the harms threatened by the listed proportion of people of color in the U.S. ii. Qualitative Benefits Such as Impacts pollutants. The commenters noted that population as a whole. The same study on Tribal Culture and Practices Congress itself listed the pollutants, found that people living within 3 miles Some commenters stated that the rather than waiting for the EPA to do so, of such power plants had an average EPA’s proposed approach ignores non- because of a difficulty which annual per capita income of $18,596, monetizable benefits. These commenters commenters argue is particular to air significantly lower than the national asserted that methylmercury toxics: ‘‘[t]he public health average. contamination threatens traditional consequences of substances which Some commenters pointed to various American Indian lifeways, including express their toxic potential only after executive orders that independently longstanding traditions of fishing and long periods of chronic exposure will direct the EPA to consider some of these fish consumption that are central to not be given sufficient weight in the factors, including Executive Order many tribes’ cultural identity and that regulatory process when they must be 12898 (February 11, 1994), which make individual tribes as distinct as balanced against the present-day costs establishes that ‘‘disproportionately different individual people. These of pollution control and its other high and adverse human health or commenters stated that for many tribes, economic consequences.’’ Leg. Hist. at environmental effects’’ of EPA decisions fishing and fish consumption are critical 8522 (S. Rep. No. 101–228 at 182). The ‘‘on minority populations and low- social practices, handed down from commenters argued that these identified income populations in the U.S. and its generation to generation. Where tribal harms from air toxics occur regardless of territories and possessions’’ are of members no longer fish due to health the source of the pollutants, and, central concern to the EPA’s decision- concerns, these fishing traditions are not therefore, there is no reason to believe making, with specific emphasis upon passed down to new generations of that Congress might have, by inserting ‘‘subsistence consumption of fish and

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00011 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 31296 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations

wildlife.’’ The commenters also pointed taken into account,’’ the Court found II.C.3 of this preamble) differently than to Executive Order 13045 (April 21, that a more detailed justification might the manner in which the EPA evaluated 1997),14 which is particularly concerned be warranted than what would suffice them in the 2016 Supplemental Finding. about ‘‘environmental health risks’’ that for a new policy. While it is true that many of the benefits may ‘‘disproportionately affect Although commenters assert that the associated with reducing emissions of children.’’ EPA has failed to provide a reasoned HAP from power plants have not been Response: Agency decisions, once basis for its action here, their real quantified, the EPA provided in the made, are not forever ‘‘carved in stone.’’ complaint with the Agency’s 2019 Proposal its reasons for concluding Nat’l Cable & Telecomms. Ass’n v. abandonment of the 2016 Supplemental that those unquantified benefits were Brand X Internet Servs., 545 U.S. 967, Finding’s ‘‘cost reasonableness test’’ and not likely to overcome the imbalance 981 (2005) (internal quotation marks ‘‘preferred approach’’ is that they between the monetized HAP benefits and citations omitted). We disagree with favored the way the Agency under that and compliance costs in the record. the commenters’ view that the EPA is approach weighed certain factors, First, as the EPA pointed out and as not permitted to determine that the including unquantified benefits, discussed below, most of the ‘‘cost reasonableness’’ approach is not impacts on tribes and tribal culture, the unquantified benefits of MATS are the correct way to consider cost in the latency and persistence of air toxics in morbidity effects associated with CAA section 112(n)(1)(A) appropriate the environment, and distributional exposure to mercury and other HAP. and necessary finding, and their view concerns and impacts. That the EPA Second, to the extent commenters have that the EPA is not permitted to re- now weighs these concerns differently— identified potential mortality outcomes evaluate the significance of the factual a weighing that is further explained such as potential cardiovascular impacts findings underpinning its 2016 below—does not mean the Agency is from mercury exposure and potential Supplemental Finding and come to a ‘‘disregarding’’ or ‘‘dismissing’’ these cancer risks from exposure to other different conclusion. D.C. Circuit and concerns. HAP, the EPA disagrees, for the reasons U.S. Supreme Court precedent, In the 2019 Proposal, the EPA clearly provided below, with the proposition including those cases cited by the stated that the unquantified HAP that significant monetized benefits commenters, support the Agency’s benefits associated with regulating would be expected from either outcome. position that it is within its authority to power plants were ‘‘significant,’’ and As the commenters acknowledged, do so, provided that the Agency’s new enumerated the impacts on human the SAB noted that IQ loss ‘‘is not the action is based on a permissible health that have been linked to mercury most potentially significant health effect interpretation of the statute and is (including neurologic, cardiovascular, associated with mercury exposure, as supported by a reasoned explanation. genotoxic, and immunotoxic effects), other neurobehavioral effects, such as In FCC v. Fox, the U.S. Supreme Court the adverse health effects associated language, memory, attention, and other with non-mercury HAP (including stated an agency’s obligation with developmental indices, are more cancer and chronic and acute health respect to changing a prior policy quite responsive to mercury exposure.’’ 80 FR disorders that implicate organ systems plainly: 75040. The Agency explained in its such as the lungs and kidneys), and 2019 Proposal that the neurobehavioral We find no basis . . . for a requirement other effects on wildlife and ecosystems. effects of mercury exposure identified that all agency change be subjected to more 84 FR 2677. Contrary to commenters’ by the SAB as more ‘‘potentially searching review. The [Administrative assertions, the EPA did not ignore these significant’’ are morbidity, not Procedure] Act mentions no such heightened standard. And our opinion in State Farm concerns but said, ‘‘The EPA mortality, outcomes. In the EPA’s neither held nor implied that every agency acknowledges the importance of these experience, the economic value of action representing a policy change must be benefits and the limitations on the avoided morbidity effects (e.g., impaired justified by reasons more substantial than Agency’s ability to monetize HAP- cognitive development, problems with those required to adopt a policy in the first specific benefits. The EPA agrees that language, abnormal social development, instance.15 such benefits are relevant to any etc.) per incident is a small fraction of In cases where an agency is changing comparison of the benefits and costs of the monetizable value of avoided its position, the Court stated that a a regulation.’’ Id. at 2677–78. Moreover, premature deaths. Further, when reasoned explanation for the new policy as the Agency pointed out in its estimating the economic value of would ordinarily ‘‘display awareness proposal, the 2011 RIA, which avoided cases of air pollution-related that it is changing position’’ and ‘‘show summarizes the factual findings and effects, the Agency has generally found that there are good reasons for the new scientific studies which form the basis that the aggregate value of the avoided policy.’’ Id. at 515. However, the Court of this action as well as the EPA’s 2016 illnesses (e.g., hospital admissions, held that the agency ‘‘need not action, discussed all of the monetized emergency department visits, cases of demonstrate . . . that the reasons for the and unquantified benefits of regulating aggravated asthma, etc.) is small as HAP from power plants, including the compared to the total value of avoided new policy are better than the reasons 19 for the old one; it suffices that the new qualitative impacts on American Indian deaths. 16 17 And the EPA does not expect that to policy is permissible under the statute, tribes, distributional impacts, and the extent the prevention of any that there are good reasons for it, and latency and persistence of the 18 premature deaths due to regulation of that the agency believes it to be better.’’ pollutant. Id. at 2678. Id. In cases where a new policy ‘‘rests In the context of this action, in which the lens we use to consider cost is based 19 See U.S. EPA 2010a: Regulatory Impact upon factual findings that contradict Analysis for the Nitrogen Oxide National Ambient those which underlay its prior policy; or on a comparison of benefits to cost, we Air Quality Standards Page 4–8 through 4–10; U.S. when its prior policy has engendered are choosing to weigh these concerns EPA. 2010b: Regulatory Impact Analysis for the serious reliance interests that must be (and particulate matter (PM) co-benefits Sulfur Dioxide National Ambient Air Quality discussed in more detail in section Standards Page 5–26 through 5–28; U.S. EPA. 2012: Regulatory Impact Analysis for the Particulate 14 Commenters cite Executive Order 13035 in Matter National Ambient Air Quality Standards their comments, but we believe this was a 16 2011 RIA at 7–40 to –49. pages 5–69; U.S. EPA. 2015: Regulatory Impact typographical error. 17 2011 RIA at 7–49 to –54. Analysis for the Ozone National Ambient Air 15 FCC v. Fox, 556 U.S. at 514. 18 2011 RIA at Chapter 4. Quality Standards. Pages 6–57 through 6–60.

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00012 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations 31297

HAP could be associated with the distribution of those benefits is. The 397, 416 (1967) (declaring that an MATS rule, the value of that effect EPA, therefore, believes it is reasonable agency, ‘‘in light of reconsideration of would be significant. With respect to to conclude that those factors to which the relevant facts and its mandate, may potential premature deaths due to the EPA previously gave significant alter its past interpretation and overturn cardiovascular impacts from mercury weight—including qualitative benefits, past administrative rulings’’); Organized exposure, as discussed further in section and distributional concerns and impacts Village of Kake v. Dept. of Agriculture, II.C.4 of this preamble, there is on minorities—will not be given the 795 F.3d 956 (9th Cir. 2015) (‘‘We do inconsistency among available studies same weight in a comparison of benefits not question that the Department was as to the degree of association between and costs for this action under CAA entitled in 2003 to give more weight to methylmercury exposure and various section 112(n)(1)(A).21 socioeconomic concerns than it had in cardiovascular system effects, including None of the information underlying 2001, even on precisely the same studies showing no association. As a the EPA’s action here constitutes new record.’’). result, based on the presently available factual findings, but rather is a As alluded to in these cases, the information, the EPA believes available reevaluation of the existing record to ‘‘reasoned basis’’ for an agency’s change evidence does not support a clear arrive at what the Agency believes to be of interpretation need not be overly characterization of the potential the better policy regarding whether complex. Even Justice Breyer, who relationship between mercury exposure regulation is ‘‘appropriate.’’ In Nat’l dissented from the FCC v. Fox majority, and cardiovascular mortality. For that Ass’n of Home Builders v. EPA, the D.C. admitted, ‘‘I recognize that sometimes reason, the EPA has not modeled risk Circuit reviewed challenges brought the ultimate explanation for a change (incidence) estimates for this health against the EPA that were similar to may have to be, ‘We now weight the endpoint and has not included benefits those concerns raised by commenters relevant considerations differently.’ ’’ associated with that endpoint in the here and found that ‘‘this kind of 556 U.S. at 550. Such change can, and analysis. With respect to potential reevaluation is well within an Agency’s often is, fueled by the basic functioning premature deaths associated with discretion.’’ 682 F.3d 1032, 1038 (D.C. of American democracy—when new inhalation exposure to non-mercury Cir. 2012) (NAHB). There, the EPA presidential administrations come into HAP, based on existing case-study reversed course on a prior policy, and office—and the courts have recognized analyses for EGUs which focus on the petitioners in that case contended that this to be a legitimate basis for a re- assessment of individual risk based on ‘‘EPA has provided no justification for weighing of priorities. See NAHB, 682 a number of conservative assumptions its decision to reverse course . . . that F.3d at 1038 (noting the ‘‘inauguration regarding exposure, the EPA anticipates is grounded in any information or of a new President and the confirmation that the mortality incidence associated experience that was not available to the of a new EPA Administrator’’ largely with these non-mercury HAP exposures Agency when it [adopted] the original provided the reasoning for the EPA’s would be low (see section II.C.3 of this rule . . . Rather, EPA merely revisited change in policy). Unlike in State Farm, preamble for additional detail).20 In old arguments that had already been where the administering agency issued sum, while the EPA recognizes the addressed as part of the original a rollback of a regulation requiring importance of unquantified benefits in a rulemaking.’’ NAHB, 682 F.3d at 1036. passive restraints in automobiles comparison against costs, the evaluation Petitioners insisted in that case that the without even mentioning airbags at all, of evidence of unquantified benefits is Agency was required to be held to a 463 U.S. at 48, 49, 51, here we acknowledge and address those factors based on qualitative information that higher standard in reversing its prior decision based on the same factual to which we are giving less weight than helps understand the likelihood and record, but the D.C. Circuit disagreed. was given in the 2016 Supplemental potential scale of those benefits, relative The Court held that FCC v. Fox Finding. Cf. Organized Village of Kake, to the monetized benefits and ‘‘foreclosed’’ petitioners’ argument, and 795 F.3d at 968 (suggesting that a policy monetized costs. These qualitative that the Agency was permitted to rely reversal could be premised upon assessments help confirm that on ‘‘a reevaluation of which policy ‘‘merely decid[ing] that [the agency] unquantified benefits do not alter the would be better in light of the facts.’’ Id. valued socioeconomic concerns more underlying conclusion that costs greatly at 1036–38. It is well settled that such highly than environmental protection’’). outweigh HAP benefits. This topic is re-weighing or re-balancing is The commenters disagree with the way discussed in more detail in section permissible. See State Farm, 463 U.S. at the Agency has now weighed the facts II.C.3 of this preamble. 57 (‘‘An agency’s view of what is in the and circumstances underlying the The other factors identified by the public interest may change, either with original appropriate and necessary commenters concern qualitative or without a change in circumstances.’’); finding and the Agency’s consideration concerns such as impacts to tribal Am. Trucking Ass’ns v. Atchison, of cost in 2016. But that does not mean cultures and the concentration of public Topeka & Santa Fe Ry. Co., 387 U.S. that the Agency has not provided a health risks occurring among certain ‘‘reasoned basis’’ for its action. population subgroups or for individuals 21 Nor does the EPA agree with the commenters Comment: Some commenters asserted living proximate to EGUs. The that Executive Orders 12898 and 13045 require a that a ‘‘more detailed justification’’ of distribution of potential health effects particular outcome in the EPA’s appropriate and the EPA’s change in policy is required may indicate more risk to some necessary finding. Executive orders recognize that agencies must weigh conflicting goals, priorities, in this case given the ‘‘serious reliance individuals than to others or more and associated costs as a necessary part of reasoned interests’’ of states, the public, and impacts to some groups like tribes than decision making. Other more recent executive industry in maintaining the appropriate others; but in a cost-benefit comparison, orders, which emphasize the environmentally and necessary determination and the the overall amount of the benefits stays responsible use and development of domestic natural resources, are also part of the policy MATS rule (citing Fox, 556 U.S. at 515; the same no matter what the calculus to consider. See, e.g., Executive Order No. Encino Motorcars, LLC v. Navarro, 136 13783, 82 FR 16093 (March 28, 2017) (directing the S. Ct. 2117 (2016)). With respect to state 20 U.S. EPA, Supplement to the Non-Hg Case EPA to review for possible reconsideration any rule and public interests, the commenters Study Chronic Inhalation Risk Assessment In that could ‘‘potentially burden the development or Support of the Appropriate and Necessary Finding use of domestically produced energy resources, pointed to the fact that the for Coal- and Oil-Fired Electric Generating Units, with particular attention to oil, natural gas, coal, implementation of MATS has led to a November 2011, EPA–452/R–11–013. and nuclear energy resources. dramatic decrease in HAP emissions

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00013 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 31298 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations

from power plants, and that the public explaining why that policy is more 3. The EPA’s Alternative Benefit-Cost has an interest in having those controls consistent with statutory language than Approach Used in the 2016 remain in place and in the continuation alternative policies,’’ Id. (internal Supplemental Finding Improperly of improvements in air quality and the citations omitted), but chided the DOL Considered Co-Benefits From Non-HAP corresponding public health and for failing to include such a justification Emissions Reductions environmental benefits. Other in its policy reversal. The 2016 Supplemental Finding First, we note that commenters raising commenters pointed to the major capital presented an alternative approach under serious reliance interests differ in at investments that regulated utilities have which the EPA made an independent least one major way from the petitioners already made to comply with MATS finding under CAA section 112(n)(1)(A) and asserted that a reversal of the 2016 in Encino Motorcars. While those petitioners faced very real impacts based on a formal benefit-cost Supplemental Finding creates analysis 22 that it was appropriate and uncertainty for the standards based on the Agency’s changed position (‘‘systemic, significant’’ changes to necessary to regulate EGUs under CAA themselves. The commenters argued section 112. See 81 FR 24427. The that these reliance interests, which they employee compensation and potential liabilities from failure to comply with formal benefit-cost analysis used in the claim depend on the maintenance of the 2016 Supplemental Finding relied on 2016 Supplemental Finding, therefore, the changed policy), the reliance interests cited by the commenters are information reported in the RIA require the EPA to provide the developed for the 2012 MATS Final heightened justification required under not upended by this final action. As we stated in the proposal, the EPA finds Rule pursuant to Executive Orders Fox and Encino Motorcars for its 12866 and 13563 and applicable statutes reversal of that finding. that its re-evaluation of the costs and benefits of regulation of HAP emissions other than the CAA (e.g., the Regulatory Response: The EPA disagrees with the from power plants will not rescind or Flexibility Act and the Unfunded commenters that the Agency is required affect the regulatory program upon Mandates Reform Act), as informed by to provide a ‘‘heightened justification’’ which the commenters rely, due to Office of Management and Budget for this action. In Fox, the U.S. Supreme 23 binding D.C. Circuit precedent (see (OMB) guidance and the EPA’s Court stated that as a general matter, no 24 section II.D of this preamble). To the Economic Guidelines. heightened scrutiny or review applies to contrary, the EPA is finalizing the The quantified benefits accounted for decisions by agencies to reverse results of the proposed RTR of MATS in in the formal benefit-cost analysis in the policies, and that policy changes need this final action. The EPA determined 2016 Supplemental Finding’s not be justified by reasons more that after compliance with MATS, the alternative approach included both HAP substantial than those required to adopt residual risks due to emissions of HAP and non-HAP air quality benefits. Based a policy in the first instance. See Fox, from the Coal- and Oil-Fired EGU on the 2011 RIA, the EPA projected the 556 U.S. at 514–15. But the Court noted source category are acceptable in quantifiable benefits of HAP reductions that ‘‘in such cases it is not that further accordance with CAA section 112, and under the rule to be $4 to $6 million in justification is demanded by the mere that there are no developments in HAP 2015.25 The RIA also identified fact of policy change; but that a emissions controls to achieve further unquantified benefits associated with reasoned explanation is needed for cost-effective reductions beyond the reducing HAP emissions from EGUs. disregarding facts and circumstances current standards. Therefore, based on that underlay or were engendered by the the results of the RTR analyses, the 22 We use the term ‘‘formal benefit-cost analysis’’ prior policy, i.e., . . . when its prior Agency is promulgating this final action to refer to an economic analysis that attempts to policy has engendered serious reliance quantify all significant consequences of an action in that maintains MATS in its current monetary terms in order to determine whether an interests that must be taken into form. action increases economic efficiency. A benefit-cost account.’’ Id. at 515. The Court Second, unlike the DOL in Encino analysis evaluates the favorable effects of policy elaborated on this principle in Encino Motorcars, the EPA has provided its actions and the associated opportunity costs of Motorcars v. Navarro, 136 S. Ct. 2117 those actions. The favorable effects are defined as reasons for changing its determination benefits. Opportunities forgone define economic (2016). There, the Court found that the that the regulation of HAP emissions costs. A formal benefit cost analysis seeks to retail automobile and truck dealership from power plants is not ‘‘appropriate.’’ determine whether the willingness to pay for an industry had relied for decades on the As explained in the proposal and in this action by those advantaged by it exceeds the Department of Labor’s (DOL) position willingness to accept the action by those preamble, the EPA believes that a disadvantaged by it. The key to performing benefit- that service advisors are exempt from consideration of costs that compares the cost analysis is the ability to measure both benefits the Fair Labor Standard Act’s overtime costs of compliance with the HAP- and costs in monetary terms so that they are pay requirements. Given this reliance specific benefits of regulation ‘‘is more comparable. Assuming all consequences can be and the impact that the DOL’s change in monetized, actions with positive net benefits (i.e., consistent with statutory language’’ than benefits exceed costs) improve economic efficiency. policy would have on the industry the 2016 Supplemental Finding’s This usage is consistent with the definition of a (citing ‘‘systemic, significant changes to ‘‘preferred approach.’’ Further, as benefit-cost analysis used in the economics the dealerships’ compensation discussed in section II.C.3 of this literature and the EPA’s Guidelines for Preparing arrangements’’ and the risk that non- preamble, we do not think the Economic Analyses. 23 U.S. OMB. 2003. Circular A–4 Guidance to conforming dealerships could face determination that regulation is Federal Agencies on Preparation of Regulatory ‘‘substantial FLSA liability’’), the Court ‘‘appropriate’’ under CAA section Analysis. Available at https://www.whitehouse.gov/ held that the DOL had not provided 112(n)(1)(A), an air toxics provision, sites/whitehouse.gov/files/omb/circulars/A4/a- good reasons for its change in policy, should primarily hinge on the monetary 4.pdf. noting that the agency ‘‘said almost benefits associated with reductions in 24 U.S. EPA. 2014. Guidelines for Preparing Economic Analyses. EPA–240–R–10–001. National nothing’’ and that it merely stated that emissions of pollutants not regulated Center for Environmental Economics, Office of exempting such employees from under CAA section 112. We believe the Policy. Washington, DC. December. Available at overtime pay was contrary to the statute explanations provided in this action https://www.epa.gov/environmental-economics/ and it believed its interpretation was fully comply with the case law’s guidelines-preparing-economic-analyses. Docket ID Item No. EPA–HQ–OAR–2009–0234–20503. reasonable. Encino Motorcars, 136 S. Ct. requirement to provide a reasoned 25 Like the 2011 RIA, all benefits and costs in this at 2126–27. The Court stated that ‘‘an explanation for our reversal of the 2016 and subsequent sections of this preamble are agency may justify its policy choice by Supplemental Finding. reported in 2007 dollars.

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00014 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations 31299

The EPA projected that the co-benefits EPA explained that unquantified conduct a study of hazards to health associated with reducing these non-HAP benefits are relevant to any comparison likely to occur from utility HAP pollutants would be substantial. Indeed, of the benefits and costs of regulation. emissions after implementation of other these projected co-benefits comprised Because unquantified benefits are, by non-HAP provisions of the CAA, and the overwhelming majority definition, not considered in monetary suggested that this provision implies (approximately 99.9 percent) of the terms, the EPA proposed that the that the EPA should evaluate non-HAP monetized benefits of MATS ($36 Administrator would evaluate the benefits of HAP regulations to see if billion to $89 billion in 2015). The evidence of unquantified benefits and they are sufficient to establish the case compliance costs of the 2012 MATS determine the extent to which they alter for HAP regulation. One commenter Final Rule were projected to be $9.6 any appropriate and necessary noted that the EPA’s approach billion in 2015.26 These compliance conclusion based on the comparison of arbitrarily excludes from consideration costs are an estimate of the increased monetized costs and benefits. a critically important set of the consequences of the EPA’s decision, expenditures in capital, fuel, and other b. Final Rule inputs by the entire power sector to namely the public health concerns at comply with MATS emissions The EPA is finalizing the the heart of the CAA. requirements, while continuing to meet determination outlined in the 2019 Response: The EPA agrees with the a given level of electricity demand. Proposal. The EPA believes that the commenters that it is critical to examine alternative approach to the 2016 the language in CAA section a. Summary of 2019 Proposal Supplemental Finding was 112(n)(1)(A), as well as the overall The EPA proposed to find that it had fundamentally flawed in applying a context of CAA section 112, in erred in the 2016 Supplemental formal cost-benefit analysis to the determining the scope of the cost Finding’s benefit-cost analysis in giving specific decision making standard consideration for the appropriate and equal weight to the air quality co- directed by CAA section 112(n)(1)(A) necessary determination. In CAA benefits projected to occur as a result of because, in the context of the section 112, Congress has a the reductions in HAP. The focus of appropriate and necessary finding, particularized focus on reducing HAP CAA section 112(n)(1)(A) is HAP doing so implied that an equal weight emissions and addressing public health emissions reductions. was given to the non-HAP co-benefit and environmental risks from those The EPA outlined in detail in the emission reductions and the HAP- emissions. In CAA section 112(n)(1)(A), 2019 Proposal that the Agency had erred specific benefits of the regulation. The Congress directs the EPA to decide in concluding in the 2016 Supplemental total cost of compliance with MATS whether regulation of EGUs is Finding that the statutory text of CAA ($9.6 billion in 2015) vastly outweighs— appropriate and necessary under CAA section 112(n)(1)(A) and the legislative by a factor of 1 thousand, or 3 orders of section 112, i.e., whether the history of CAA section 112 more magnitude—the monetized HAP deployment of specific CAA provisions generally supported the position that it benefits of the rule ($4 to $6 million in targeted at reducing HAP emissions was reasonable to give equal weight to 2015). In these circumstances, to give from the EGU sector is warranted. The co-benefits in a CAA section equal weight to the monetized PM2.5 co- EPA believes that it cannot answer this 112(n)(1)(A) appropriate and necessary benefits would permit those benefits to question by pointing to benefits that are finding. 81 FR 24439. The EPA become the driver of the regulatory overwhelmingly attributable to explained in the 2019 Proposal that, determination, which the EPA believes reductions in an entirely different set of because the vast majority of the would not be appropriate for the reasons pollutants not targeted by CAA section estimated monetized benefits in the stated in the proposal and set forth 112. The EPA believes that it is illogical 2011 RIA that were estimated to result below. for the Agency to make a determination, from MATS are associated with c. Comments and Responses informed by a study of what hazards reductions in fine particulate matter remain after implementation of other Comment: Many commenters argued (PM ) precursor emissions, the EPA CAA programs, that regulation under 2.5 that the EPA’s proposed approach to had erred in the 2016 Supplemental CAA section 112, which is expressly considering co-benefits in the CAA Finding by giving equal weight to non- designed to deal with HAP emissions, is section 112(n)(1)(A) appropriate and HAP co-benefits in making the ‘‘appropriate’’ principally on the basis necessary determination is not appropriate and necessary of criteria pollutant impacts. consistent with the statute. The determination. As the 2019 Proposal The EPA believes that relying almost commenters believe that basic observed, Congress, in the National exclusively on benefits accredited to principles of statutory construction do Ambient Air Quality Standards reductions in pollutants not targeted by not allow the EPA to read CAA section (NAAQS) program, established a CAA section 112 is particularly 112(n)(1)(A) only in isolation. The rigorous system for setting standards of inappropriate given that those other commenters asserted that the EPA has acceptable levels of criteria air pollutants are already comprehensively not explained why CAA section pollutants requisite to protect public regulated under other CAA provisions, 112(n)(1)(A)’s reference to regulation of health with an adequate margin of such as those applying to the NAAQS. EGUs allows the Agency to disregard a safety, and by state, regional, and As the EPA outlined in the 2019 portion of the consequences of its Proposal, the determination that it is not national rulemakings establishing decision. One commenter noted that the appropriate to give equal weight to non- control measures to meet those levels. The EPA did acknowledge the language in the Senate Report on the HAP co-benefits in making the importance of unquantified benefits in 1990 amendments to CAA section 112, appropriate and necessary the 2019 Proposal, but also pointed out which directs the EPA to consider the determination is further supported by the limitations of the Agency’s ability to co-benefits of HAP regulation, is the the fact that Congress established a monetize HAP-specific benefits. The closest specific indication of rigorous system for setting standards of congressional intent for interpreting acceptable levels of criteria air 26 See Table 3–5 of the RIA: https:// CAA section 112(n). The commenter pollutants and provided a www3.epa.gov/ttn/ecas/docs/ria/utilities_ria_final- also pointed to the portion of CAA comprehensive framework directing the mats_2011-12.pdf. section 112(n) that requires the EPA to implementation of those standards in

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00015 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 31300 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations

order to address the health and section, if the Administrator finds such contrary, the Court said that the proper environmental impacts associated with regulation is appropriate and necessary treatment of co-benefits is ‘‘a point we those pollutants. See, e.g., 42 U.S.C. after considering the results of the study need not address.’’ 135 S.Ct. at 2711. 7409; 7410; 7501; 7502; 7505a; 7506; required by this subparagraph.’’ The text Additionally, commenters seem to 7506a; 7507; 7509; 7509a; 7511; 7511a; on its face suggests that Congress mistake the EPA’s position (see, e.g., 7511b; 7511c; 7511d; 7511e; 7511f; wanted the Administrator’s appropriate Environmental Protection Network 7512; 7512a; 7513; 7513a; 7513b; 7514; and necessary determination to be (EPN) comment at 25 (April 17, 2019) and 7515. The vast majority of the focused on the health hazards related to (Docket ID Item No. EPA–HQ–OAR– monetized benefits in the 2011 RIA that HAP emissions and the potential 2018–0794–2261) (referring to ‘‘EPA’s were estimated to result from MATS are benefits of avoiding those hazards by crabbed claim that it can focus only on associated with reductions in PM2.5 reducing HAP emissions. While the reduction of ‘HAP emissions—without precursor emissions, principally provision in one sense does even considering reductions in non- nitrogen oxides (NOX) and sulfur acknowledge the existence of co- HAP pollutants’).’’ See also States and dioxide (SO2). NOX, SO2, and PM2.5 are benefits—i.e., by referencing the Local Governments comment at 35–36 already addressed by a multitude of potential for ancillary reductions of (April 17, 2019) (Docket ID Item No. statutory provisions governing levels of HAP emissions by way of CAA EPA–HQ–OAR–2018–0794–1175) (‘‘In these pollutants, including the NAAQS provisions targeting other pollutants—it proposing to exclude consideration of provisions that require the EPA to set does not follow from this that any [co-benefits], EPA misinterprets and standards for criteria pollutants ancillary reductions of criteria misapplies the Supreme Court’s requisite to protect public health with pollutants that may be projected to directive in Michigan.’’)). The an adequate margin of safety, and by result from the regulation of EGU HAP commenters essentially argue that the state, regional, and national rulemakings emissions should, therefore, play a part language in Michigan requires the EPA establishing control measures to meet in the Administrator’s consideration to review ‘‘all the relevant factors,’’ those levels. under CAA section 112(n)(1)(A) including co-benefits. As described at The 2016 Supplemental Finding whether the regulation of EGUs is length in the 2019 Proposal and other pointed to CAA section 112(n)(1)(A)’s ‘‘appropriate and necessary.’’ To the parts of this section of this preamble, directive to ‘‘perform a study of the contrary, the statutory direction to the EPA is considering what hazards to public health reasonably consider whether it is appropriate and significance co-benefits have for its anticipated to occur as a result of necessary to regulate HAP after criteria determination under CAA section emissions by electric utility steam pollutants have been addressed by the 112(n)(1)(A)—but we are concluding generating units of [HAP] after CAA’s other requirements suggests that that the finding must be justified imposition of the requirements of [the it is not proper for the co-benefits of overwhelmingly by the HAP benefits CAA],’’ and noted that the requirement further criteria pollutant reductions to due to the statutory structure. to consider co-benefit reduction of HAP provide the dominant justification for Comment: Some commenters argued resulting from other CAA programs an affirmative CAA section 112(n)(1)(A) that existing case law, beyond the highlighted Congress’ understanding determination. Certainly, Congress’ Michigan decision, supports inclusion that programs targeted at reducing non- instruction to the EPA that it study HAP of indirect benefits into an agency’s HAP pollutants can and do result in the effects under CAA section 112 after benefit-cost analysis. A commenter reduction of HAP emissions. Id. The implementation of other CAA quoted the D.C. Circuit’s statement in finding also noted that the Senate provisions cuts against any suggestion American Trucking Ass’ns v. EPA that Report on CAA section 112(d)(2) that such benefits should be given equal the EPA must consider both the direct recognized that MACT standards would consideration in a CAA section and indirect effects of pollutants, rather have the collateral benefit of controlling 112(n)(1)(A) determination. than only ‘‘half of a substance’s health criteria pollutants. Id. However, these Comment: Several commenters argued effects.’’ 175 F.3d 1027, 1051–53 (D.C. statements acknowledging that that the EPA’s proposed approach, of Cir. 1999), rev’d on other grounds sub reductions in HAP can have the not providing consideration to co- nom. Whitman v. Am. Trucking Ass’ns, collateral benefit of reducing non-HAP benefits equal to the consideration Inc., 531 U.S. 457 (2001). The emissions and vice versa, provides no provided to the benefits specific to HAP commenter also cited a Fifth Circuit support for the proposition that any reductions, takes a too-narrow approach case in which the Court held that the such co-benefits should be considered that conflicts with Michigan. EPA had to consider the indirect safety on equal footing as the HAP-specific Commenters pointed out that the Court harm that could result from the use of benefits when the Agency makes its found that CAA section 112(n) tells the substitute, non-asbestos brakes when finding under CAA section 112(n)(1)(A). EPA to undertake a ‘‘broad and all- attempting to ban asbestos-based brakes The study referenced in CAA section encompassing’’ review of ‘‘all the under the Toxic Substances Control Act. 112(n)(1)(A) specifically focuses on the relevant factors.’’ 135 S. Ct. at 2707. Corrosion Proof Fittings v. EPA, 947 hazards to public health that will Commenters argued that if the Court F.2d 1202, 1225 (5th Cir. 1991). A few reasonably occur as a result of HAP read ‘‘appropriate’’ to be a ‘‘broad and commenters also noted the D.C. emissions, not harmful emissions in all-encompassing term,’’ then the EPA Circuit’s favorable treatment of the general. (‘‘The Administrator shall cannot excise relevant factors from EPA’s consideration of co-benefits in perform a study of the hazards to public consideration. Commenters also stated regulating HAP from boilers, process health reasonably anticipated to occur that the Court, in instructing the EPA to heaters, and incinerators in U.S. Sugar as a result of emissions by electric consider cost, appeared to adopt a broad Corp. v. EPA, 830 F.3d 579, 591, 625 utility steam generating units of reading of the word ‘‘cost,’’ including (D.C. Cir. 2016). pollutants listed under subsection (b) of ‘‘more than the expense of complying Response: As explained elsewhere in this section after imposition of the with regulations; any disadvantage this preamble, the EPA is interpreting requirements of this chapter.’’) could be termed a cost.’’ 137 S. Ct. at and applying the statutory directive to According to that section, ‘‘[t]he 2707. make an appropriate and necessary Administrator shall regulate electric Response: Nothing in the Michigan determination under CAA section utility steam generating units under this decision decides this issue. To the 112(n)(1)(A) and determining what role

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00016 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations 31301

consideration of co-benefits should play designed to address. In fact, the EPA Response: The EPA developed the in making that determination. None of believes that it would be arbitrary and 2011 RIA for the 2012 MATS Final Rule the case law the commenters cite capricious to do so. See Motor Vehicle pursuant to Executive Orders 12866 and pertains to CAA section 112(n)(1)(A), Mfrs. Ass’n v. State Farm Mut. Auto. 13563, as well as certain other and, therefore, the case law is not Ins. Co., 463 U.S. 29, 43 (1983) applicable statutes, as informed by OMB directly relevant to this action. (‘‘Normally, an agency rule would be guidance and the EPA’s Economic As explained in the 2019 Proposal arbitrary and capricious if the agency Guidelines. It is true that, in this action, and in this preamble, the EPA believes has relied on factors which Congress has the EPA is drawing on information that it would be inconsistent with the not intended it to consider.’’). generated in that RIA in order to make statute and with case law to base the The EPA is not turning a blind eye to the determination required under CAA appropriate and necessary finding on a the reasonably predictable section 112(n)(1)(A) concerning whether monetized benefit estimate that is consequences of MATS. The 2011 RIA regulation of EGUs under CAA section almost exclusively attributable to appropriately details the magnitude of 112 is appropriate. How costs are to be reductions of non-HAP pollutants. the PM2.5-related co-benefits in the form considered in making the Further, the CAA sets out a specific of avoided premature deaths, hospital congressionally-directed CAA section regulatory scheme for the PM pollutants admissions, emergency department 112(n)(1)(A) determination, however, is in question, the NAAQS, and as a first visits and asthma attacks, among other not governed independent from principle the EPA believes those endpoints. However, CAA section statutory requirements, by preexisting regulations, not CAA section 112, 112(n)(1)(A) requires a threshold OMB or EPA guidelines, nor could it be. should be the primary method by which determination of whether any regulation Furthermore, for the many reasons the Agency targets those pollutants. of EGUs under CAA section 112 is explained elsewhere in this preamble Comment: Several commenters argued ‘‘appropriate and necessary.’’ The EPA and in the 2019 Proposal, the CAA that the EPA’s approach of giving less believes that this inquiry must be section 112(n)(1)(A) determination is weight to co-benefits in the appropriate focused primarily on the risks posed by governed by the particular statutory and necessary determination is the pollutants targeted by CAA section provision at issue, and, therefore, is fundamentally arbitrary. The 112, i.e., HAP emissions. The gross distinct from any other CAA action. commenters pointed out that the PM2.5 disparity between monetized costs and In the context of conducting the CAA emission reductions are a direct result HAP benefits, which should be the section 112(n)(1)(A) determination, the of HAP emissions controls, and that primary focus of the Administrator’s there is no way to reduce the HAP determination in CAA section EPA finds it is not only appropriate but emissions without reducing PM 112(n)(1)(A), is so great as to make it indeed, necessary for the EPA to emissions. Some commenters asserted inappropriate to form the basis of the interpret and apply the particular that excluding some benefits from the necessary statutory finding. While the provision of CAA section 112(n)(1)(A), appropriate and necessary Agency acknowledges that PM co- which as mentioned earlier specifically determination creates a biased analysis. benefits are substantial, the Agency cites to HAP listed under section 112(b) One commenter argued that the EPA’s cannot rely on PM co-benefits to of the CAA. To be valid, the EPA’s approach is arbitrary and contrary to supplant the primary factors Congress analytical approach to that provision Michigan and other U.S. Supreme Court directed the Administrator to consider. must recognize Congress’ particular precedent because it ‘‘fai[ls] to consider Comment: Several commenters concern about risks associated with [such] an important aspect of the asserted that the EPA’s approach to HAP and the benefits that would accrue problem.’’ Michigan, 135 S. Ct. at 2707 considering co-benefits under the CAA from reducing those risks. OMB and (quoting State Farm, 463 U.S. at 53). section 112(n)(1)(A) analysis was EPA guidance outline regulatory Response: The EPA acknowledges the inappropriate because it is principles that agencies are encouraged existence and importance of these co- unprecedented in the EPA’s regulatory to follow to the extent permissible benefits. However, when the EPA is practice and contrary to OMB and EPA under law. These guidance documents, comparing benefits to costs as a required policy. Commenters asserted that co- and the standard economic principles prerequisite to regulation, it is critical to benefits are universally accepted as an reflected in them, are not necessarily examine the particular statutory important tool in regulatory economics informative regarding how Congress provision that is being implemented. and economic planning. Commenters intended the EPA to make the CAA That statutory provision may limit the quoted OMB Circular A–4 as directing section 112(n)(1)(A) determination, nor relevance of certain costs and benefits— agencies in conducting RIAs to ‘‘look should they be read to override statutory e.g., serve to establish that any benefits beyond the direct benefits and direct text and structure that, as explained attributable to the ancillary reduction of costs of your rulemaking and consider earlier in this preamble, requires a focus pollutant emissions that are not the any important ancillary benefits and on a limited set of costs and benefits. focus of the provision at issue are not countervailing risks.’’ The commenters Although an analysis of all reasonably ‘‘an important aspect of the problem’’ also identified the EPA’s ‘‘Guidelines anticipated benefits and costs in that Congress is seeking to address. As for Preparing Economic Analyses’’ that accordance with generally recognized noted in the 2019 Proposal and in states: ‘‘An economic analysis of benefit-cost analysis practices earlier responses to comments, in CAA regulatory or policy options should (including extending analytic efforts to section 112(n)(1)(A), Congress directs present all identifiable costs and ancillary impacts in a balanced manner the EPA to decide whether regulation of benefits that are incremental to the across both benefits and costs) is EGUs is appropriate and necessary regulation or policy under appropriate for informing the public under CAA section 112; the EPA consideration. These should include about the potential effects of any believes that it is not appropriate to directly intended effects and associated regulatory action, as well as for answer this question in the affirmative costs, as well as ancillary (or co-) complying with the requirements of by pointing to benefits that are benefits and costs.’’ Commenters also Executive Order 12866, it does not overwhelmingly attributable to cited to previous clean air rules where follow that equal consideration of all reductions in an entirely different set of the EPA has afforded co-benefits equal benefits and costs, including co- pollutants that CAA section 112 is not weight in cost-benefit analyses. benefits, is warranted, or even

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00017 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 31302 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations

permissible, for the specific statutory necessary. As noted previously in this benefits of controlling non-mercury provision requiring the EPA to make an preamble and the 2019 Proposal, the metal HAP were unquantified. Again, appropriate and necessary finding vast majority of estimated monetized the vast majority of estimated monetized called for under CAA section benefits resulting from MATS are benefits resulting from MATS are 112(n)(1)(A). associated with reductions in PM2.5 associated with reductions in premature Comment: Some commenters asserted precursor emissions, principally NOX mortality resulting from emissions that the EPA’s 2019 Proposal and SO2. Both NOX and SO2 are criteria reductions of PM precursors and not erroneously suggests that CAA sections pollutants in their own right and are from metal HAP or even direct PM. 110 and 112 must be treated as mutually already addressed by the numerous Comment: Several commenters exclusive authorities for reducing the statutory provisions governing criteria asserted that the EPA has not explained public health impacts of PM emissions. pollutants. In interpreting and applying what weight is given to co-benefits, or Commenters argued that there is no CAA section 112(n)(1)(A), we believe it how the EPA chose that standard, aside basis to ignore the benefits of reducing is important to acknowledge that the from saying that the weight is less than pollutants merely because they are also CAA has established numerous robust what is given to HAP-specific benefits. subject to regulation under state and avenues for minimizing PM-precursor One commenter noted that the EPA federal implementation plans approved emissions to a level that is requisite to essentially claims that co-benefits to implement the NAAQS. One protect public health with an adequate cannot affect the appropriate and commenter noted that the existence of margin of safety. Because other CAA necessary determination unless other CAA provisions that deal with programs are already in place to ensure quantified HAP benefits are criteria pollutant emissions likely reductions in criteria pollutants to the ‘‘moderately commensurate’’ with indicates Congress’ deep concern about level requisite to protect public health compliance costs, but the EPA does not the health and environmental risks they with an adequate margin of safety, the provide any clarity on the point at pose. One commenter argued that there EPA believes that it is not reasonable to which HAP benefits would be is no legal support for the idea that CAA point to criteria pollutant co-benefits as ‘‘moderately commensurate’’ to allow section 110 or 112 requires exclusivity; the primary benefit to justify regulation the EPA to rely on co-benefits. the EPA is not required to pick one of EGUs under a provision of the CAA Response: The Administrator has avenue through which it can impact PM that authorizes such regulation only concluded that the following procedure emissions. The commenter noted that where the Administrator determines provides the appropriate method under many CAA provisions can address PM, that it is ‘‘appropriate and necessary’’ to which the EPA should proceed to such as those for interstate transport and do so.27 determine whether it is appropriate and regional haze, and the EPA itself has With respect to one commenter’s necessary to regulate EGUs under CAA encouraged states in their assertion that the EPA’s approach was section 112(n)(1)(A). First, the EPA implementation planning to consider particularly unfounded given that many compares the monetized costs of selecting controls that will minimize metal HAP are emitted as PM, the EPA regulation against the subset of HAP emissions of multiple pollutants. agrees that most non-mercury metal benefits that could be monetized. Here, Another commenter acknowledged that HAP are emitted as PM. In fact, the EPA those costs are disproportionate to the the EPA does not argue that the other established an emission standard for monetized benefits, by three orders of provisions should be the exclusive filterable PM in the 2012 MATS Final magnitude. That does not demonstrate vehicle for addressing criteria Rule that serves as a surrogate for the ‘‘appropriate and necessary.’’ Second, pollutants, but this commenter asserted non-mercury metal HAP (recognizing the EPA considers whether unquantified that the 2019 Proposal did not explain that controls for PM are also effective for HAP benefits may alter that outcome. how criteria pollutant reductions could the non-mercury metal HAP). However, For the reasons proposed in February be realized more effectively by some the fact that the non-mercury metal HAP 2019 and further discussed in this final other legal mechanism and did not are emitted in a solid particulate form action, the EPA determines they do not. claim that criteria pollutants have been does not mean that the EPA should give Third, the EPA considers whether it is fully controlled through those other equal weight to the benefits from appropriate, notwithstanding the above, programs. One commenter also argued removal of all PM. As described in the to determine that it is ‘‘appropriate and that the EPA’s proposal is particularly 2011 RIA for the 2012 MATS Final Rule, necessary’’ to regulate EGUs under CAA unfounded because many metal HAP PM2.5 benefits result from emissions section 112(n)(1)(A) out of consideration are emitted as PM. reductions of SO2 (1,330,000 tons), NOX for the PM co-benefits that result from Response: The EPA disagrees with the (46,000 tons), carbonaceous PM2.5 (6,100 such regulation. For the reasons commenters. The EPA’s discussion of tons), and crustal PM2.5 (39,000 tons). proposed in February 2019 and set forth co-benefits, and the impropriety of Control of directly-emitted filterable PM in this final action, on the record before giving them equal weight to HAP- for purposes of controlling non-mercury the Agency, it is not appropriate to do specific benefits within the context of metal HAP constituted approximately 5 so. the appropriate and necessary percent of the total PM2.5 health co- Here, almost the entirety of monetized determination, is based on an benefits of the rule. Based on analysis of benefits (about 99.9 percent) of MATS interpretation of CAA section available data, the EPA estimates that 112(n)(1)(A), a provision enacted by non-mercury metal HAP represent, at EPA developed ratios of non-mercury metal and Congress to address the unique situation most, 0.8 percent of this directly emitted filterable PM emissions for use in estimating facing EGUs. We have limited our filterable PM.28 The actual HAP-related emissions from coal- and oil-fired EGUs without analysis to the specifically tailored current non-mercury metal emissions data. These ratios were determined by dividing the fuel-specific provision of CAA section 112(n)(1)(A), 27 A number of commenters raised this same issue averages of the 2010 MATS Information Collection in which Congress recognized that EGUs and made this same point. See, e.g., Docket ID Item Request (ICR) non-mercury metals data, combined would face regulation under numerous Nos. EPA–HQ–OAR–2018–0794–1135, –1178, by control technique where possible, by the parts of the CAA and chose to ask the –1189, –1190. filterable PM emissions data. The ratios represent 28 As mentioned in the Emission Factor the amount of non-mercury metals present in EPA to consider whether further Development for RTR Risk Modeling Dataset for filterable PM. For more detail, see memorandum regulation of EGUs under CAA section Coal- and Oil-fired EGUs memorandum (Docket ID titled Non-mercury Metals Content of Filterable 112 would be appropriate and Item No. EPA–HQ–OAR–2018–0794–0010), the Particulate Matter in the docket for this action.

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00018 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations 31303

reflected in the RIA were derived from the reduced-form technique used to ignore co-benefits while giving full non-HAP co-benefits. Had the HAP- relate PM2.5 emission precursors to the weight to indirect compliance costs. specific benefits of MATS been closer to number and value of PM2.5 adverse Response: The EPA disagrees with the the costs of regulation, a different health effects; and the approach used to commenters that co-benefits and the question might have arisen as to assign a dollar value to adverse health types of compliance costs that the whether the Administrator could find effects. The Agency has separately noted commenters consider ‘‘indirect’’ must that co-benefits legally form part of the that, in general, it is more confident in be given comparable treatment within justification for determination that the size of the risks we estimate from this action. As discussed throughout regulation of EGUs under CAA section simulated PM2.5 concentrations that this section, the EPA believes that it is 112(d) is appropriate and necessary. The coincide with the bulk of the observed inappropriate to rely, as did the EPA does not need to, and does not, PM concentrations in the alternative, benefit-cost approach in the determine whether that additional step epidemiological studies that are used to 2016 Supplemental Finding, almost would be appropriate in this factual estimate the benefits. Likewise, the exclusively on benefits accredited to scenario given that the monetized and Agency is less confident in the risk reductions in pollutants not targeted by CAA section 112 when those other unquantified HAP-specific benefits do estimated from simulated PM2.5 not come close to a level that would concentrations that fall below the bulk pollutants are already extensively support the prior determination. Under of the observed data in these studies.29 regulated under other CAA provisions. Additionally, unlike benefits, which the interpretation of CAA section Furthermore, when setting the 2012 PM can be disaggregated into benefits 112(n)(1)(A) that the EPA adopts in this NAAQS, the Administrator attributable to reduction in HAP and co- action, HAP benefits, as compared to acknowledged greater uncertainty in benefits attributable to reduction in non- costs, must be the primary question in specifying the ‘‘magnitude and HAP pollutants, costs cannot similarly making the ‘appropriate and necessary’ significance’’ of PM-related health risks be disaggregated. There is no analogous determination. While the Administrator at PM concentrations below the could consider air quality benefits other distinction with respect to compliance NAAQS. As noted in the preamble to costs and, thus, nothing in the statute than HAP-specific benefits in the CAA the 2012 PM NAAQS final rule, in the section 112(n)(1)(A) context, that directs the EPA to partition context of selecting an alternative compliance costs into direct and consideration of these co-benefits could NAAQS, the ‘‘EPA concludes that it is permissibly play only, at most, a indirect (or ancillary) costs, or that not appropriate to place as much supports the view that such a marginal role in that determination, confidence in the magnitude and given that the CAA has assigned partitioning would be appropriate. significance of the associations over the From an economic perspective, MATS regulation of criteria pollutants to other lower percentiles of the distribution in was a consequential rulemaking that provisions in title I of the CAA, each study as at and around the long- was expected to induce changes in both specifically the NAAQS regime term mean concentration.’’ (78 FR 3154, electricity and fuel markets beyond the pursuant to CAA sections 107–110, January 15, 2013). impacts on affected coal- and oil-fired which requires the EPA to determine Comment: Some commenters argued EGUs. The policy case examined in the what standards for the ambient that the EPA is inappropriately giving 2011 RIA introduced the requirements concentration of PM are necessary to full weight to the consideration of of MATS as constraints on affected protect human health. Here, to the indirect costs of regulating EGUs while EGUs, which resulted in new extent that the alternative approach set simultaneously giving less than equal projections of power sector outcomes forth within the 2016 Supplemental weight to co-benefits. One commenter under MATS. These compliance costs Finding was legally grounded in co- argued that comparing direct and are an estimate of the increased benefits, the massive disparity between indirect costs to only the ‘‘direct’’ expenditures in capital, fuel, labor, and co-benefits and HAP benefits on this benefits associated with HAP reductions other inputs by the entire power sector record would mean that that alternative is not an apples-to-apples comparison. to comply with MATS emissions approach clearly elevated co-benefits Some commenters stated that the EPA is requirements, while continuing to meet beyond their permissible role. including not only compliance costs a given level of electricity demand. If the Administrator were to consider incurred by the sources regulated under These costs were summarized in Table the size of the PM2.5-related co-benefits MATS, but also costs incurred by other 3–16 of the 2011 RIA.30 in deciding whether regulating EGUs power plants that are not regulated The commenters do not attempt to under CAA section 112(d) is appropriate under MATS due to the effects on the present an alternative analysis under and necessary, he should also consider power sector of regulated sources’ which the EPA would assess what they taking into account key assumptions investing in pollution abatement term ‘‘indirect costs.’’ To focus on the affecting the size and distribution of technologies or taking other steps to projected impact of MATS on only these co-benefits and potential reduce emissions. The commenter affected entities would produce an uncertainty surrounding them. In the argued that the EPA does not explain incomplete estimate of the entire cost of past, the EPA has highlighted a number why it is appropriate to discount or complying with the rule and, thus, lead of these assumptions as having to an inappropriate consideration of the costs of the 2012 MATS Final Rule. The particularly significant effect on 29 The Federal Register document for the 2012 estimates of PM-related benefits, PM NAAQS indicates that ‘‘[i]n considering this costs termed ‘‘indirect costs’’ by including assumptions about: The additional population level information, the commenters are neither ancillary or causal relationship between PM Administrator recognizes that, in general, the incidental costs; these costs are an confidence in the magnitude and significance of an integral part of the compliance costs exposure and the risk of adverse health association identified in a study is strongest at and effects; the shape of the concentration- around the long-term mean concentration for the air that are attributable to expected changes response relationship for long-term quality distribution, as this represents the part of 30 exposure-related PM2.5 and the risk of the distribution in which the data in any given The EPA estimated the impacts of MATS on premature death; the toxicity of study are generally most concentrated. She also oil-fired units and costs associated with monitoring, recognizes that the degree of confidence decreases recordkeeping, and reporting in separate analyses, individual PM2.5 particle components; as one moves towards the lower part of the which are summarized in Chapter 3 and Appendix the levels of future PM2.5; the validity of distribution.’’ 3A of the 2011 RIA.

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00019 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 31304 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations

to production behavior in the sector in the public health benefits from reducing associated with avoiding premature order to minimize the cost of complying HAP emissions (see also Gwinn, et al., death, and in particular, potential with MATS. Furthermore, an evaluation 2011,31 and Fann, Wesson, and Hubbell, cancer risks.34 As part of the 2012 of the costs borne solely by the owners 2016 32 for a detailed discussion of the MATS Final Rule, the EPA modeled the of EGUs subject to MATS would need complexities associated with estimating maximum individual risk (MIR) to account for the ability of owners of the benefits of reducing emissions of air associated with non-mercury HAP these EGUs to recoup their increased toxics). These obstacles include gaps in including arsenic, hexavalent expenditures through higher electricity toxicological data, uncertainties in chromium, nickel, and hydrogen prices; otherwise, an estimate of the extrapolating results from high-dose chloride for a subset of 16 EGUs. MIR costs of MATS borne by the owners of animal experiments and worker studies is the ‘‘maximum individual risk’’ those EGUs (i.e., their economic to estimate human effects at lower experienced by the most highly exposed incidence) would be an overestimate. doses, limited monitoring data, individual living in proximity to the However, if the EPA was to only difficulties in tracking diseases such as source, presuming continuous exposure account for the economic incidence for cancer that have long latency periods, for 70 years. The analysis found that the owners of EGUs, the costs borne by the and insufficient economic research to one oil-fired EGU studied had a lifetime consumers of electricity from these support the valuation of the health cancer risk of 20-in-1 million, and that higher prices would be ignored, which impacts often associated with exposure none of the remaining 15 coal-fired EGU the EPA finds inappropriate. Therefore, to individual HAP. facilities posed a lifetime risk of cancer the EPA determined it was appropriate The EPA fully acknowledges the for the maximally exposed individual to account for all of the costs that may existence and importance of the exceeding 8-in-1 million, with most be incurred as a result of the rule that unquantified benefits. The EPA facilities posing a risk of equal to, or less could be reasonably estimated, explained in the 2019 Proposal reasons than, 1-in-1 million. These risks are recognizing that these expenditures why the EPA has determined that the significantly below the levels defined by would ultimately be borne either by unquantified benefits are unlikely to the EPA as being the presumptive upper electricity consumers or electricity overcome the significant difference limit of acceptable risk (i.e., 1-in-10 producers, rather than limiting our (which, the EPA notes again, is a thousand). While that analysis did not consideration of costs to just those difference of three orders of magnitude) separately estimate the number of new borne by a subset of producers or between the monetized HAP-specific cases of HAP-attributable cancer among consumers. benefits and compliance costs of the each year, the size of the MIR implies Comment: Some commenters asserted MATS rule. This is also further that the number of new cases would that the EPA has failed to explain how discussed in section II.C.2 of this likely be very small. The EPA’s it has given any meaningful preamble. As noted there, many of the evaluation of evidence of unquantified consideration in its benefit-cost HAP-related effects that were benefits is based on qualitative comparison to the numerous health unquantified in the 2011 RIA consist of information that helps understand the effects of reducing HAP emissions that morbidity effects in humans. The EPA’s likelihood and potential scale of those the EPA has not quantified. A few methods estimating the economic value benefits, relative to the monetized commenters asserted that the non- of avoided health effects values benefits and monetized costs. These monetized benefits of the rule mortality effects significantly more than qualitative assessments help confirm encompass virtually all the HAP avoided illnesses (e.g., hospital that unquantified benefits do not alter reductions that the rule yields. One admissions, emergency department the underlying conclusions that costs commenter argued that the EPA has visits, cases of aggravated asthma, greatly outweigh HAP benefits. only given ‘‘lip service’’ to these etc.).33 Hence, valuing HAP-related Comment: Several commenters benefits, but not any discernible weight morbidity outcomes would not likely pointed out that the EPA’s 2019 in reaching the conclusion that result in estimated economic values Proposal relies on undefined terms such regulating EGUs under CAA section 112 similar to those attributed to avoiding as ‘‘moderately commensurate,’’ ‘‘gross is not appropriate and necessary. premature deaths. disparity,’’ and ‘‘significant difference,’’ Further, the commenter asserted that the Commenters raised the possibility which are not statutory terms and do EPA has offered no support or that there could be unquantified HAP- not appear in prior regulatory actions explanation for the assertion that the related benefits of mortality effects, unquantified benefits are not sufficient based on the comments the EPA associated with MATS. Without to overcome the difference between the believes the most significant are explanation of what these terms mean, monetized benefits and the costs of the commenters asserted that the public MATS. 31 Gwinn, M.R., et al., 2011. Meeting Report: did not receive adequate notice so that Response: The 2011 RIA attempted to Estimating the Benefits of Reducing Hazardous Air they could provide meaningful account for all the monetized and Pollutants—Summary of 2009 Workshop and comments on the proposal. Commenters Future Considerations. Environmental Health said the 2019 Proposal leaves the public unquantified benefits of the rule, and Perspectives, 119(1): 125–130. the EPA’s benefit-cost analysis in the 32 Fann N., Wesson K., and Hubbell B (2016), in the dark as to what data and RIA does not discount the existence or Characterizing the confluence of air pollution risks methodology the EPA relies on to importance of the unquantified benefits in the United States. Air Qual Atmos Health 9:293. determine that the costs of regulating Available at https://doi.org/10.1007/s11869-015- power plants under CAA section 112 of reducing HAP emissions. However, in 0340-9. this final action, the EPA has 33 See U.S. EPA. 2010a: Regulatory Impact ‘‘grossly outweigh’’ the hazardous air determined that it is reasonable to Analysis for the Nitrogen Oxide National Ambient pollution benefits. One commenter evaluate unquantified benefits Air Quality Standards Page 4–8 through 4–10; U.S. asserted that the failure to define these EPA. 2010b: Regulatory Impact Analysis for the terms and outline the EPA’s analytical separately in the comparison of benefits Sulfur Dioxide National Ambient Air Quality and costs for this action under CAA Standards Page 5–26 through 5–28; U.S. EPA. 2012: section 112(n)(1)(A). Regulatory Impact Analysis for the Particulate 34 See sections II.C.2 and II.C.4 of this preamble The EPA explained in the 2011 RIA Matter National Ambient Air Quality Standards for the EPA’s response to commenters’ assertions pages 5–69; U.S. EPA. 2015: Regulatory Impact regarding potential mortality effects due to that there are significant obstacles to Analysis for the Ozone National Ambient Air methylmercury exposure and cardiovascular successfully quantifying and monetizing Quality Standards. Pages 6–57 through 6–60. impacts.

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00020 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations 31305

methodology has rendered this action in 4. It Is Reasonable To Continue To Rely EPA used the results from the 2011 RIA violation of CAA section 307(d). on the Original 2011 Regulatory Cost- for the updated comparison, as this RIA Response: The EPA believes that the Benefit Data Comparison as Part of a contained the best available information language used in its 2019 Proposal and CAA Section 112(n)(1)(A) Assessment of on the projected costs, benefits, and impacts of the MATS rule at the time final actions is reasonable and Costs and Benefits the Agency was making its regulatory understandable and is consistent with a. Summary of 2019 Proposal decision to establish CAA section 112(d) legal standards that have been As discussed above, in the 2016 emissions standards. previously upheld in litigation Supplemental Finding, the EPA challenges. For example, in the Entergy considered an alternative approach to b. Final Rule decision the U.S. Supreme Court upheld considering cost as part of the The EPA is finalizing the the EPA’s use of a ‘‘wholly appropriate and necessary finding that determination outlined in the 2019 disproportionate’’ standard. 556 U.S. at was based on a benefit-cost analysis Proposal. The EPA believes that the 224 (‘‘[I]t is also not reasonable to originally performed as part of the 2011 approach to the formal benefit-cost interpret Section 1326(b) as requiring RIA for the 2012 MATS Final Rule. This analysis presented in the 2011 RIA use of technology whose cost is wholly analysis summarized the EPA’s contains the best available information disproportionate to the environmental projected estimates of annualized on the projected costs, benefits, and benefit to be gains’’) (internal quotation benefits, costs, and net benefits of the impacts of the MATS rule at the time removed). Further, as recognized in the MATS rule in 2015. The 2011 RIA the Agency was making its regulatory 2016 Supplemental Finding, CAA considered costs, quantified HAP decision to establish CAA section 112(d) section 112(n)(1)(A) and the Michigan benefits, unquantified HAP benefits, emissions standards. The EPA maintains that, based upon an decision give broad discretion to the and non-HAP co-benefits and concluded evaluation of the information in the Administrator to apply his expert that aggregated monetized benefits ($37 record, even if the Agency were to judgment in considering cost in order to to $90 billion each year) exceeded the costs of compliance ($9.6 billion) by 3 perform new analysis to estimate the determine whether it is appropriate and to 9 times. The EPA, therefore, benefit and cost impacts of MATS, the necessary to regulate HAP emissions concluded in the 2016 Supplemental results are unlikely to materially alter from EGUs. See 81 FR 24428. CAA Finding’s alternative approach that the the general conclusions of the analysis, section 112(n)(1)(A) requires that ‘‘the RIA’s benefit-cost analysis supported its with small benefits associated with the Administrator shall regulate [EGUs] . . . affirmation of the prior appropriate and targeted quantified HAP benefits and if the Administrator finds such necessary finding under CAA section compliance costs and would not alter regulation is appropriate and 112(n)(1)(A). the final determination herein. necessary.’’ The Michigan Court The 2019 Proposal also used the c. Comments and Responses explicitly acknowledged the discretion estimates from the 2011 RIA to address held by the Administrator: ‘‘[i]t will be costs in the context of a CAA section Comment: Some commenters asserted up to the Agency to decide (as always, 112(n)(1)(A) appropriate and necessary that the EPA has failed to comply with within the limits of reasonable finding but concluded that the basic principles of administrative law interpretation) how to account for cost.’’ alternative approach in the 2016 by failing to develop an adequate factual 135 S. Ct. at 2711. As explained in the Supplemental Finding had improperly record in basing its cost-benefit prior response and in other places in weighed the non-HAP co-benefits comparison on the data contained in the 2011 RIA, as opposed to gathering the this preamble, the EPA has concluded, estimates reported in the 2011 RIA. body of information relevant to these as a result of our qualitative evaluation Specifically, the EPA concluded that the issues that has since become available. of evidence, that unquantified benefits Agency’s previous equal weighting of These commenters asserted that any cannot reasonably be expected to be the PM2.5 co-benefits projected to occur as a result of the reductions in HAP consideration of the appropriate and comparable to the cost of regulation or necessary finding must consider new to meaningfully redress the gross emissions was inappropriate given that the focus of CAA section 112(n)(1)(A) is information on what the benefits and disparity between that cost and the costs of regulating EGUs would be if the monetized HAP benefits. The on the HAP emissions reductions themselves. Upon reconsideration, the question were revisited in light of commenters take issue with some of the EPA proposed to determine that it current knowledge, not as the facts were terminology used in the 2019 Proposal, would be illogical for the Agency to thought to be 8 years in the past. but given the discretion afforded to the decide that regulation under CAA Response: The EPA agrees with the Administrator by CAA section section 112, which is expressly commenters that courts have required 112(n)(1)(A), as acknowledged by the designed to deal with HAP, could be administrative agencies to address U.S. Supreme Court, we believe this justified primarily based on the non- ‘‘newly acquired data in a reasonable preamble outlines a reasonable and HAP pollutant impacts of these fashion,’’ but depending on the fitting approach to Congress’ open- regulations. In the 2019 Proposal, the circumstances, agencies are not always ended instruction to the Administrator EPA provided an updated comparison required to rely on updated data when to determine whether a regulation of of costs and targeted pollutant benefits engaged in decision-making. American EGUs is ‘‘appropriate and necessary.’’ (i.e., HAP benefits) in a memorandum to Iron & Steel Inst. v. EPA, 115 F.3d 979, The EPA further believes that, in a the proposed rulemaking docket.35 The 1007 (D.C. Cir. 1997). The EPA context where costs outweigh maintains that its use of benefit and cost monetized HAP-specific benefits by 35 See Compliance Cost, HAP Benefits, and information from the 2011 RIA is three orders of magnitude, the meaning Ancillary Co-Pollutant Benefits for ‘‘National reasonable in this context. Emission Standards for Hazardous Air Pollutants: To determine whether an agency and relevance of terms such as ‘‘gross Coal-and Oil-Fired Electric Utility Steam reasonably addressed updated data, disparity’’ and ‘‘significant difference’’ Generating Units—Reconsideration of courts may look to the statutory are self-evident. Supplemental Finding and Residual Risk and Technology Review’’ (Docket ID Item No. EPA–HQ– mandate to the Agency. NRDC v. OAR–2018–0794–0007). Herrington, 786 F.2d 1355 (D.C. Cir.

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00021 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 31306 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations

1985). Under the statutory structure of the rulemaking (e.g., in setting the in the 2011 RIA ($9.6 billion versus $4 CAA section 112, the CAA section emission standards or in the RIA). The to $6 million annually in 2015) would 112(n)(1)(A) finding is a preliminary Court emphasized that the Agency itself be materially different under any re- determination that is made significantly had not relied upon these rationales at analysis.36 Several commenters pointed before the CAA section 112(d) standards the finding stage. 135 S. Ct. 2710–11 to independent analyses that provided would be promulgated. The suggestion (citing SEC v. Chenery Corp., 318 U.S. three estimates of the actual costs of by some commenters that the EPA is 80, 87 (1943)). However, the Court left MATS. While none of these estimates required to conduct a new analysis that open the possibility that the economic can be precisely compared against the attempts to estimate the actual costs analyses the Agency had already EPA ex ante estimates because they use incurred through compliance with the conducted could suffice to satisfy its different cost metrics and dollar years, final CAA section 112(d) standards is, obligation to consider costs as part of the independent analyses indicate that, thus, not consistent with the statute. the appropriate finding. Id. at 2711. if actual costs were to be estimated in The 2016 Supplemental Finding There is nothing in the operative a manner consistent with the EPA’s similarly declined to conduct new statutory language here that is akin to 2011 RIA estimates, the compliance analysis before reaffirming the wording that courts have found to costs expenditures would still likely be appropriate and necessary require an agency to incorporate in the billions of dollars. determination, arguing that this was an updated information. See Sierra Club v. First, a 2015 analysis by Andover appropriate approach to the problem EPA, 671 F.3d 955 (9th Cir. 2012) Technology Partners referred to by because that determination is a (directing the EPA to rely on updated commenters estimated that the actual threshold question under the statute. 81 data when approving nonattainment cost of compliance in the initial years of FR 24432 (2016 Supplemental Finding). state implementation plans (SIPs) implementation was approximately $2 We also note that in 2012, the EPA because CAA section 172(c)(3) requires billion per year.37 38 The second study interpreted CAA section 112(n)(1)(A) as SIPs to include ‘‘comprehensive, referred to by commenters was a study not obligating the Agency to update its accurate, current inventory of actual performed by M.J. Bradley & Associates data, and we maintain that emissions’’); see also City of Las Vegas interpretation here. That interpretation v. Lujan, 891 F.2d 927 (D.C. Cir. 1989) (MJB&A) using information from the (holding that the Secretary of the U.S. Energy Information is consistent with the text and structure 39 of CAA section 112(n)(1)(A), which Interior could not disregard available Administration. MJB&A estimated that focuses on an expressly required study scientific information because the MATS-regulated facilities incurred total that evaluates hazards to public health. Endangered Species Act required the capital expenditures on environmental When the EPA reaffirmed the 2000 ‘‘best scientific and commercial data retrofits of $4.45 billion, an estimate appropriate and necessary finding in available’’). that does not include ongoing operating In addition to looking at the statutory 2012, it explained that although it was and maintenance expenditures. Finally, language, courts also often examine the choosing to undertake an updated as documented in a letter to the EPA impact any updated data would have analysis of the public health risks and cited by several commenters, the had on the agency’s decision. Catawba associated with EGU HAP emissions, Edison Electric Institute estimated that County v. EPA, 571 F.3d 20, 45 (D.C. doing so was ‘‘not required.’’ 77 FR the power sector incurred total Cir. 2009) (upholding the EPA’s 9304, 9310 (February 16, 2012). The compliance costs of more than $18 designations for the NAAQS because EPA argued at the time that the billion, including both capital and ‘‘EPA dealt with the newly acquired continued existence of the appropriate data in a reasonable fashion by 36 and necessary finding in 2012 was The EPA’s April 15, 2020, finalization of the explaining why it would not have subcategorization of Eastern Bituminous Coal warranted by the analysis undertaken in changed the designations’’); see also Refuse-Fired EGUs could alter the benefits and 1998 and summarized in the 2000 Eastern Carolinas Broadcasting v. FCC, costs of MATS. However, given that such appropriate and necessary finding. Id. subcategorization will affect only six units, we 762 F.2d 95, 98 (D.C. Cir. 1985) think it is reasonable to expect that any changes to Both the statute and the Michigan (upholding FCC’s determination in light the 2011 RIA’s projected cost and benefits as a decision support the EPA’s reliance on of the Commission’s failure to utilize result of the potential subcategorization would not the cost estimates from the 2011 RIA. updated data because it was a ‘‘harmless materially affect the EPA’s conclusion that First, any cost analysis included in an compliance costs of MATS disproportionately error in light of the ultimate rationale’’). outweigh the HAP benefits associated with the ‘‘initial decision to regulate,’’ Michigan, According to the commenters, costs of standards. 135 S. Ct. at 2709, must precede any MATS compliance have been lower 37 Declaration of James E. Staudt, Ph.D., CFA, at regulations flowing out of that decision. than the EPA estimated in 2011 and the 3, White Stallion Energy Center v. EPA, No. 12– Therefore, in considering the costs of EPA has not accounted for more recent 1100 (D.C. Cir., December 24, 2015). Also available at Docket ID Item No. EPA–HQ–OAR–2009–0234– compliance as part of its appropriate studies of quantified HAP benefits. 20549. and necessary finding, it is reasonable However, even if the EPA updated its 38 In addition to the 2015 study, Andover for the EPA to look at what types of cost analysis, there is no reason to believe Technology Partners produced two other analyses information, such as the 2011 RIA cost that the new data and analysis would in 2017 and 2019, respectively, that estimated the estimates, would be available at this change the overall conclusion of the ongoing costs of MATS. The 2017 report estimated that the total annual operating cost for MATS- threshold stage. In addition, nothing in 2011 analysis that costs outweighed the related environmental controls was about $620 the Michigan decision precludes the quantified benefit attributed to million, an estimate that does not include ongoing EPA’s use of the existing cost reduction in HAP emissions. payments for installed environmental capital. The information in the record in addressing However, while it is challenging to 2019 report estimates the total annual ongoing incremental costs of MATS to be about $200 the Agency’s obligation on remand to produce rigorous retrospective estimates million; again, this estimate does not include consider cost as part of the appropriate of the benefits and costs of MATS, it is ongoing MATS-related capital payment. The 2017 and necessary finding. In Michigan, the possible to demonstrate, using publicly report is available in Docket ID Item No. EPA–HQ– Court rejected arguments that it could available information, that there is no OAR–2018–0794–0794. The 2019 report is available in Docket ID Item No. EPA–HQ–OAR–2018–0794– conclude that the Agency had properly reason to believe that the relative 1175. considered cost based on the Agency’s difference between compliance costs 39 Available in Docket ID Item No. EPA–HQ– consideration of costs in other stages of and quantified HAP benefits projected OAR–2018–0794–1145.

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00022 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations 31307

operations and maintenance costs.40 change in the relative magnitude of was developed at the time when the While these retrospective cost estimates costs and HAP-related benefits. In Agency reaffirmed the appropriate and are developed from bases that are satisfaction of the requirements of necessary finding and established CAA dissimilar from one another and, in OMB’s Circular A–4, Section 3 of the section 112(d) standards for EGUs. particular, from how the EPA developed memorandum, Compliance Cost, HAP Additionally, as discussed in another the prospective cost estimates in the Benefits, and Ancillary Co-Pollutant comment response in this section, even 2011 RIA, it is evident that the Benefits, that accompanies this final if actual compliance costs are lower independent analyses each indicate that action presents all reasonably than the EPA projected in the 2011 RIA, the industry costs of MATS are of a anticipated costs and benefits arising the costs are still likely to be at least an similar order of magnitude and in the out of the MATS rule, including those order of magnitude greater than the billions of dollars. arising out of co-benefits. monetized HAP benefits. At the same time, the quantified Comment: Commenters said that the Comment: Other commenters rejected mercury-related benefits would still compliance cost estimates underlying the argument that actual utility sector likely be in the millions of dollars and the 2019 Proposal are several times compliance costs for MATS have been not substantially more than what was higher than actual costs because the less than predicted in 2011. One estimated when the rule was finalized. projections in the 2011 RIA assumed commenter said that utilities have spent Table 3–4 of the 2011 RIA shows that that MATS would require the less on retrofitting power plants by the EPA estimated that MATS would installation of additional fabric filters, simply closing plants to avoid installing reduce mercury emissions from MATS- scrubber upgrades, and electrostatic costly controls. However, the regulated units about 20 tons in 2015 precipitator upgrades that were commenter also claimed that the utility (from 27 to 7 tons). According to recent subsequently not required. sector’s avoided MATS compliance EPA estimates, mercury emissions from Additionally, the commenters suggested costs did not simply disappear; they MATS-regulated units decreased by the EPA’s analysis erred because the were translated into costs borne by the about 25 tons from 2010 (pre-MATS) to projected price of natural gas was too former employees of retired coal-fired 2017 (from 29 to 4 tons).41 Even if the low in the 2011 RIA. Commenters said plants, by coal workers who have lost 25-ton decrease in mercury emissions that what they characterized as their jobs, and by the communities of from 2010 to 2017 is entirely attributed substantial inaccuracies of the 2011 RIA those displaced workers. Commenters to MATS (which would be a very strong projections render these projections an said that the 2019 Proposal continues to assumption given other economic and inappropriate basis for the proposed treat these MATS-driven ‘‘costs’’ as regulatory factors that influenced the comparison of the costs and benefits. irrelevant when considering the trajectory of mercury emissions Response: The EPA disagrees with the regulatory impacts, but the commenters downward during this period), the commenters that the entire economic said that the EPA must add these quantified mercury-related benefits are analysis that the EPA performed in the regulatory costs to its analysis as likely to be not much greater than the 2011 RIA is invalid simply because of required by Michigan. The commenter estimates in the 2011 RIA, and certainly an asserted discrepancy between cited data indicating an individual’s job would continue to be at least an order modeling projections and actual loss has a direct correlation with of magnitude smaller than the actual outcomes. See, e.g., EME Homer City adverse health outcomes. costs of MATS. Generation, L.P. v. EPA, 795 F.3d 118, Response: The 2011 RIA provided Similarly, as discussed in more detail 135–36 (D.C. Cir. 2015) (‘‘We will not estimates of employment changes for in sections II.C.2 and II.C.3 of this invalidate EPA’s predictions solely the regulated power sector and for the preamble, we would expect that the because there might be discrepancies air pollution control sector, including unquantified HAP-related benefits of between those predictions and the real estimates of employment impacts from MATS would not meaningfully redress world. That possibility is inherent in the changes in fuel demand from EGUs. the large disparity between monetized enterprise of prediction. The best model However, examining localized costs and monetized HAP benefits might predict that the Nationals will employment impacts that may arise estimated in the 2011 RIA. Lastly, win the World Series in 2015. If that from MATS compliance actions is whether the co-benefits that MATS does not happen, you can’t necessarily outside of the scope of this action. The achieved are larger or smaller than fault the model.’’). The EPA used the commenter asserts that the cost of the estimated in the 2011 RIA is not a best available data and modeling rule will result in lost income or central consideration in the EPA’s information, in accordance with employment that will, in turn, result in appropriate and necessary finding, as Executive Order 12866 and the EPA’s negative health impacts. The EPA discussed previously in section II.C.3 of economic guidelines, and provided the disagrees that this point is relevant to this preamble.42 The net result of this public with the opportunity to comment the appropriate and necessary finding. inquiry is that we believe that if the EPA on all aspects of its analysis in Comment: Commenters highlighted were to perform retrospective analysis developing the 2011 RIA. that the industry has already incurred of the impacts of MATS for the purposes The independent analyses cited by costs to implement MATS and cannot of the appropriate and necessary several commenters find that a variety recover these costs except through rate determination, the results of that of control technology costs have shown recovery and similar mechanisms. analysis would not lead to any material to be lower than the EPA’s projection Commenters argued that finalization of from the 2011 RIA. However, the a reconsideration of the appropriate and 40 Available in Docket ID Item No. EPA–HQ– suggestion that important components necessary finding under CAA section OAR–2018–0794–2267. of the actual compliance cost of MATS 112(n)(1)(A) should be based on an 41 https://www3.epa.gov/airmarkets/progress/ are lower than the Agency’s projections analysis of ongoing and future costs reports/index.html. 42 As previously discussed, section 112(n) of the does not alter the Agency’s weighed against ongoing and future CAA requires the EPA to make a finding as to determination that the analysis in the benefits, as opposed to considering past whether regulation of EGUs is ‘‘appropriate and 2011 RIA represents the best and most costs and benefits. If the EPA considers necessary’’ following consideration of hazards to comprehensive estimate of the cost of past costs that have already been public health reasonably anticipated to result from EGU emissions of HAP listed in CAA section compliance with MATS available to the incurred by the industry to comply with 112(b). EPA for use in this finding, because it MATS in connection with the proposed

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00023 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 31308 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations

rule, the Agency must consider whether 112(c)(9) requesting that coal-fired EGUs (2011) 46 which characterize those past costs might weigh in favor of be removed from the CAA section 112(c) methylmercury-related effects. These maintaining or affirming the 2016 List of Categories of Major and Area two articles concluded that Supplemental Finding. Sources, and that the EPA denied this methylmercury is both directly linked to Response: A previous response in this petition on several grounds.43 The acute myocardial infarction and section explains why the EPA’s use of EPA’s position on denial of this petition intermediary impacts that contribute to the benefit and cost estimates from the has not changed. myocardial infarction risk. They also 2011 RIA is reasonable. Additionally, Comment: Commenters stated that discussed a host of uncertainties with respect to the suggestion that the since the revised consideration of associated with methylmercury EPA estimate future costs and benefits weighing costs and benefits as part of a cardiovascular effects. flowing from this action, section II.D of CAA section 112(n)(1)(A) finding hinges Rice et al. (2010) evaluated the this preamble explains that the EPA’s on the estimation of HAP reduction benefits of a 10-percent reduction in revised determination that regulation of benefits, the EPA must make a better methylmercury exposure for U.S. HAP emissions from EGUs under CAA effort to monetize all HAP reduction populations (reflecting IQ loss and section 112 is not appropriate and benefits. These commenters asserted presumed mortality impacts). The study necessary will not remove EGUs from that new research suggests that the EPA used a probabilistic approach to address the CAA section 112(c) list of sources, underestimated the benefits associated confidence in a causal association and the previously established MATS with HAP reductions across several between methylmercury and heart rule will remain in place. As a result, effects. Specific criticisms of the EPA attacks. Importantly, they state ‘‘we there will be no changes in future HAP benefit estimation focused view the evidence for causal compliance expenditures or emissions primarily on methylmercury 44 and interpretation as relatively weak.’’ They under MATS as a result of the revised included: (1) Failure to quantify use a subjectively defined probability of determination under CAA section cardiovascular effects; (2) criticism of one-third that the association between 112(n)(1)(A). the approach used in modeling the IQ methylmercury and cardiovascular Comment: Commenters said that loss endpoint; (3) failure to consider effects is causal, acknowledging that the many utilities that expended resources other neurological endpoints besides IQ strength of the association was to comply with MATS are subject to loss; (4) failure to consider additional ‘‘modest.’’ The Rice et al. (2010) ongoing rate reviews by public utility health effects besides neurological and estimates are also sensitive to commissions regarding recovery of cardiovascular impacts; and (5) failure assumptions regarding the coefficient MATS-associated costs. Some utilities to model the full range of fish linking hair mercury to heart attack and expressed concerns that, if MATS or the consumption pathways related to the timing of the exposure-response appropriate and necessary finding is mercury emissions from EGUs. relationship. rescinded, whether through EPA action Response: After reviewing the The Roman et al. (2011) paper was a or as a result of judicial review of a additional peer-reviewed studies on workshop report from a panel convened reversal of the 2016 Supplemental health effects attributable to mercury to assess the potential for developing a Finding, stakeholders will intervene in that were submitted in the comments, concentration-response function for the rate cases before public utility the EPA concludes that the approach to cardiovascular effect from commissions, arguing that utilities’ assessing quantified and unquantified methylmercury exposure. The report investments in the MATS-required methylmercury benefits in the 2011 recommended that the EPA develop a pollution controls were imprudent and RIA, while subject to uncertainty, new dose-response relationship for should no longer be recoverable through remains valid. We address the major cardiovascular-related methylmercury their approved rates. Because of this effects. However, the study also reports reasoning, the commenters said the EPA criticisms across the five major categories of comments below. the results of a literature review that should consider the impacts on recovery yield a very small number of in vitro or of sunk costs jeopardized by a reversal i. Failure To Quantify Cardiovascular animal studies; the review characterized of the appropriate and necessary finding Effects the strength of the epidemiological in its benefit-cost analysis. Commenters cited several studies studies that assessed clinically Response: Section II.D of this significant endpoints as being preamble explains that the EPA’s regarding the linkage between methylmercury concentrations in blood ‘‘moderate.’’ The Roman et al. (2011) revised determination that regulation of review also mentions uncertainty as to EGUs under CAA section 112 is not and tissue samples and cardiovascular health. Some of the studies cited in the which exposure metric (including the appropriate and necessary will not timing of exposure and appropriate bio- remove EGUs from the CAA section comments were available to the EPA at the time of the 2011 RIA, while others marker) would provide the most robust 112(c) list of sources, and the previously statistical outcome in modeling established MATS rule will remain in were not. The former category includes Rice et al. (2010) 45 and Roman et al. cardiovascular effects. place. As a result, the EPA does not In the 2012 MATS Final Rule, the anticipate that the ability of utilities to 43 EPA also addressed comments on the recover MATS-related expenditures will 84 FR 2679–2680. 44 Additional comments also addressed the linkage between methylmercury be jeopardized as a result of this action. modeling of non-mercury HAP in the context of the exposure and cardiovascular effects. Even if MATS were to be rescinded, a appropriate and necessary risk assessment (as One of the references cited as part of the number of states have mercury rules opposed to the benefits analysis), with these EPA response was Mozaffarian et al. that would continue to mandate the use comments focusing on claims that EPA had failed to appropriately include adjustment factors of mercury controls. The EPA is addressing individual-variability and limitations in Environmental Science & Technology, 44(13): committed to working with states that using the census block-centroid approach to 5216–5224. are interested in developing their own capturing risk for the most exposed individual. 46 Roman, H.A., et al. (2011). Evaluation of the HAP-specific requirements. The EPA’s These comments are addressed in the RTC cardiovascular effects of methylmercury exposures: document. Current evidence supports development of a dose- proposal noted that, in 2011, the Utility 45 Rice, G.E., et al. (2010). A Probabilistic response function for regulatory benefits analysis. Air Regulatory Group (UARG) submitted Characterization of the Health Benefits of Reducing Environmental Health Perspectives, 119(5): 607– a petition pursuant to CAA section Methyl Mercury Intake in the United States. 614.

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00024 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations 31309

(2011), which evaluated health exposure through fish consumption in function. In addition, Axelrad et al. outcomes from two large cohorts of men the midst of the positive cardiovascular (2007) also obtained a new modeled and women in the U.S. and showed no impacts associated with fish estimate for IQ loss for the Faroes data evidence of a relationship between consumption. However, based on the from the study authors based on mercury exposure and increased information available in the existing structural equation modeling involving cardiovascular disease risk.47 This study record and material submitted during underlying neurological endpoints. And also evaluated multiple coronary heart the public comment period, the EPA finally, Axelrad et al. (2007) also used disease subtypes and concluded that believes available evidence does not a sophisticated hierarchical random- mercury exposure was not associated support a clear characterization of the effects model that can consider study-to- with the risk of nonfatal myocardial potential relationship between mercury study and endpoint-to-endpoint infarction or fatal coronary heart exposure and cardiovascular effects at variability in modeling the endpoint. disease. Based on the available scientific this time. This does not preclude the When considered in aggregate, these literature at the time of the MATS rule, possibility that later scientific work may details regarding study design the Agency concluded that there was provide more clarity as to the existence associated with Axelrad et al. (2007) inconsistency among available studies or absence of an association. lead the EPA to conclude that the effect as to the association between Further, current research is also estimate obtained from this particular methylmercury exposure and various insufficient to support modeling of the study is well supported by the cardiovascular system effects. cardiovascular mortality endpoint with underlying evidence and continues to In the second category of newer a sufficient degree of confidence for be appropriate for modeling IQ loss literature, commenters referenced the inclusion in an EPA benefits analysis benefits related to methylmercury Genchi et al. (2017) 48 review article that due to (1) questions regarding overall exposure. summarizes the methylmercury- causality and uncertainty in specifying cardiovascular literature but does not iii. Failure To Consider Other the dose-response relationship required Neurological Endpoints Besides IQ Loss report dose-response parameters. The (including the form and paper cites studies from 2002–2007 parameterization of the function) and (2) The third broad category of criticism looking at cardiovascular-related effects uncertainty in modeling the prospective related to the 2011 RIA estimation of (e.g., heart rate variability, myocardial bio-markers (e.g., hair mercury) required benefits was that the EPA failed to infarction, atherosclerosis, in part due to questions regarding the consider other neurological endpoints hypertension, etc.) for a range of temporal aspects of the exposure- besides IQ loss in modeling benefits. populations, some U.S. and some non- response relationship. Specifically, commenters asserted that U.S. The article recommends pre-existing literature 53 and more development of a dose-response ii. Criticism of the Approach Used in recent data have revealed a suite of function for methylmercury exposure Modeling the IQ Loss Endpoint more sensitive neurodevelopmental and myocardial infarctions for The second category of criticism effects than IQ loss. For example, one regulatory benefits analysis, but does related to the 2011 RIA estimation of recent study (Patel et al., 2019) 54 not provide specific recommendations benefits involves the approach used in referenced in the comments suggests an regarding which studies, effect estimates modeling IQ loss, specifically the effect association between methylmercury or functional forms to use. The authors estimate used in modeling this exposure and behavioral problems also acknowledge the need ‘‘to improve endpoint. Commenters pointed out that (specifically anxiety), even at relatively the characterization of the potential in modeling IQ loss, two studies, low prenatal exposure levels. Another linkage between methylmercury Bellanger et al. (2013) 50 and Trasande et study, Masley et al. (2012) 55 cited by exposure and the risk of cardiovascular al. (2005),51 employ effect estimates commenters concludes that cognitive disease.’’ Commenters also cited Giang significantly larger than the effect effects of methylmercury on adults are and Selin (2016) 49 as support for their estimate utilized by the EPA in the 2011 substantial enough to negate beneficial argument that the monetized benefits of RIA, which was obtained from Axelrad effects of omega-3 fatty acids among reducing mercury is greater than the et al. (2007).52 In responding to these adults who consume large amounts of EPA estimates in the proposal. This comments, the EPA notes that both of some types of fish. Finally, commenters study also acknowledges that the these alternate studies (Bellanger et al., pointed to new research (Julvez et al., 56 relevant literature (through 2016) is 2013 and Trasande et al., 2005) utilized 2013) which suggests that some relatively small and inconsistent with data from one of the three key datasets individuals might be genetically respect to the association between (Faroes study) in characterizing the susceptible to the neurological effects of methylmercury exposure and relationship between methylmercury methylmercury and that null groups cardiovascular disease. The study notes exposure and IQ loss. By contrast, which do not include these individuals that all of the literature discusses the Axelrad et al. (2007) uses data from all could mask significant impacts among challenges associated with teasing out three key studies (Faroes, Seychelles, 53 National Research Council, The Toxicological any adverse effects of methylmercury and New Zealand) in fitting their Effects of Methylmercury, 2000. https:// www.nap.edu/catalog/9899/toxicological-effects-of- 47 Mozaffarian, D.; Shi, P.; Morris, J.S.; 50 Bellanger, D., et al. (23 authors), Economic methylmercury, p. 310. Spiegelman, D.; Grandjean, P.; Siscovick, D.S.; benefits of methylmercury exposure control in 54 Patel, N.B.; Xu, Y.; McCandless, L.C.; Chen, A.; Willett, W.C.; Rimm, E.B. Mercury exposure and Europe: Monetary value of neurotoxicity prevention. Yolton, K.; Braun, J.; . . . Lanphear, B.P. (2019). risk of cardiovascular disease in two U.S. cohorts. Environmental Health, 2013, 12:3. Very low-level prenatal mercury exposure and N Engl J Med, 2011, 364, 1116–1125. 51 Trasande, L.; Landrigan, P.; Schechter, C. behaviors in children: The HOME Study. 48 Genchi, G.; Sinicropi, M.S.; Carocci, A.; Lauria, Public Health and Economic Consequences of Environmental health: A global access science G.; Catalano, A. Mercury Exposure and Heart Methyl Mercury Toxicity to the Developing Brain. source, 18(1), 4. doi:10.1186/s12940-018-0443-5. Diseases. Int. J. Environ. Res. Public Health, 2017, Environmental Health Perspectives, Vol 113, No 5, 55 Masley, S.C.; Masley, L.V.; Gualtieri, T.: Effect 14, 74. https://doi.org/10.3390/ijerph14010074. May 2005. https://doi.org/10.1289/ehp.7743. of mercury levels & seafood intake on cognitive 49 Giang, A.; Selin, N. Benefits of mercury 52 Axelrad, D.; Bellinger, D.; Ryan, L.; Woodruff, function in middle-aged adults. Integrative controls for the United States. Proceedings of the T. Dose-Response relationship of Prenatal Mercury Medicine, 11:32–40, 2012. National Academy of Sciences, Vol 113, No. 2, Exposure and IQ: An Integrative Analysis of 56 Julvez, J. and Grandjean, P. Genetic January 12, 2016. https://doi.org/10.1073/ Epidemiologic Data. Environmental Health susceptibility to methylmercury developmental pnas.1514395113. Perspectives, Vol 115, No 4, April 2007. neurotoxicity matters. Front Genet, 4: 278, 2013.

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00025 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 31310 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations

genetically susceptible within the larger population would need to be specified benefits by focusing the benefits study group. in order to increase overall confidence analysis on self-caught (recreational) Taking these comments in order, in modeling cognitive-related benefits at freshwater fish. Specifically, regarding the potential for modeling the representative population-level for commenters pointed to Trasande et al. additional neurological endpoints, this subgroup. (2005) as an example of an assessment including behavioral problems (e.g., Regarding the potential that certain that, while also modeling benefits anxiety), the EPA notes that the cited individuals could be genetically associated with controlling mercury study (Patel et al., 2019) is equivocal in susceptible to the neurological effects of emissions from U.S. power plants, more its findings, with the authors stating that methylmercury and that, consequently, fully considers exposure to they ‘‘did not find a consistent these individuals may not be fully methylmercury, including the general association between very low-level covered by existing studies consumption of commercial fish by the prenatal mercury exposure and behavior characterizing neurodevelopmental U.S. population. The Trasande et al. problem scores in children, but [they] effects of methylmercury, the EPA (2005) study employs general linear did find some evidence of an acknowledges this as a possibility. apportionment (based on estimates of association between very low-level However, the study cited by U.S. EGU emissions relative to global mercury exposure during early commenters (Julvez et al., 2013) does emissions) to estimate the fraction of pregnancy and parent-reported anxiety not provide effect estimates for these methylmercury in U.S. freshwater and scores in children.’’ The authors note potentially at-risk subgroups, which coastal fish associated with U.S. EGU that the association of low-level prevents quantitative analysis of risk emissions. A similar calculation is used mercury exposure with behavioral and associated dollar-benefits associated to estimate the fraction of problems, including anxiety, deserves with mercury-exposure in these methylmercury in non-U.S. sourced further scrutiny. The EPA concludes subgroups. commercial fish associated with U.S. that we are not yet at the point where iv. Failure To Consider Additional EGU emissions. They then apportion we can reliably model the effects of low- Health Effects Besides Neurological and their estimate of total IQ loss for level mercury exposure on children’s Cardiovascular Impacts children in the U.S. (assumed to come behavior, including anxiety. Regarding the potential for the Commenters pointed to the potential completely from fish consumption) to beneficial cognitive effects of omega-3 for methylmercury exposure to be U.S. EGU-sourced mercury versus other fatty acids in adults (resulting from fish associated with a range of additional sources. Similarly, commenters have consumption) to be partially negated by adverse health effects (besides also cited Giang and Selin (2016) as coexistent methylmercury exposure, the neurological and cardiovascular), another example of a study that EPA recognizes conceptually that this including cancer (leukemia and liver) attempts to generate a more complete could occur. However, it is important to and possible effects on the reproductive, picture of methylmercury benefits note that the effects of methylmercury hematological, endocrine (diabetes), and associated with controlling U.S. EGU on omega-3 fatty acid intake and immune systems. The EPA notes the mercury emissions, including exposures associated benefits were seen only for distinction between evidence-based associated with commercial fish the subset of the population with support for specific health effects consumption. Notably, the Giang et al. relatively elevated consumption of (potentially even including support for (2016) study uses a more sophisticated larger fish (i.e., more than 3–4 servings causal associations should it exist) and modeling approach (compared with a month, Masley et al., 2012). Modeling the ability to reliably model those health Trasande et al., 2005), to project benefits-related changes in fish endpoints quantitatively. In referencing potential benefits associated with MATS consumption typically focuses on the the above health endpoints, commenters within the United States out to 2050, general consumer rather than attempting referred to a range of study data which including application of global mercury to model benefits for a specific subset of can be used as evidence for an deposition modeling covering specific that population which can be association, including elucidation of regions associated with commercial challenging to enumerate (i.e., the potential toxicity pathways. fishing. The authors note that greater subgroup of those consuming relatively In response to these comments, the than 90 percent of U.S. commercial fish elevated levels of higher-trophic level EPA notes that in order to model a consumption, and the majority of U.S. fish)—that level of more refined health effect within a defined mercury intake, comes from marine and subgroup modeling is often reserved for population as part of a benefits analysis, estuarine sources, particularly from the scenario-based risk assessments, where high-confidence concentration-response Pacific and Atlantic Ocean basins. population enumeration is not the functions linked to clearly defined Regarding the assertion that the EPA focus. For that reason, data on how biometrics (which can themselves be should have used methodologies similar methylmercury could obscure the simulated at the population-exposure to those cited in these studies to benefits of omega-3 fatty acid intake (for level) are required. At this time, as incorporate consideration of commercial a specific higher large-fish-consuming noted earlier, with the exception of IQ fish consumption (linked to U.S. EGU segment of the population) would have loss in children, the EPA does not mercury emissions) in its benefits less utility in the context of a benefits believe research is currently sufficient analysis, the EPA again reiterates the analysis aimed at the more generalized to support quantitative assessment of importance of including only those fish-consuming population. In addition, any of these additional endpoints in the consumption pathways that can be the EPA would note potential context of a benefits analysis involving modeled with a reasonable degree of challenges in modeling this kind of mercury (accessed through a fish- confidence. Both of the studies cited trade-off related to fish consumption, consumption pathway). employ broad-scale simplifying since not only would levels of assumptions in order to link changes in methylmercury and omega-3 fatty acids v. Failure To Model the Full Range of U.S. EGU mercury emissions to need to be characterized for a broad Fish Consumption Pathways Related to potential changes in the concentration range of fish species; in addition, the Mercury Emissions From EGUs of methylmercury in commercial fish, specific mix of those types of fish A number of commenters stated that which Giang et al. (2016) suggest is consumed by the high-consuming study the EPA underestimated IQ-related responsible for the vast majority of fish-

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00026 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations 31311

related methylmercury exposure in the As described in section II.B of this Accordingly, the Court vacated the 2005 U.S. Specifically, as noted earlier, the preamble, in 2005, the EPA reversed the Delisting Rule. Trasande et al. (2005) study links U.S. 2000 determination that regulation of Based on the D.C. Circuit’s holding in EGU emissions (as a fraction of total HAP emissions from EGUs under CAA New Jersey, the EPA proposed that global emissions) to methylmercury section 112 was appropriate and finalization of the reversal of the 2016 concentrations in commercially and necessary. At that time, the EPA Supplemental Finding, much like the recreational fish consumed by the U.S. justified its decision to delist EGUs 2005 Delisting Rule’s reversal of the population. With the Giang et al. (2016) because it ‘‘reasonably interprets section 2000 appropriate and necessary study, the authors utilize U.S. EGU 112(n)(1)(A) as providing it authority to determination, would not have the deposition (as a fraction of total) in remove coal- and oil-fired units from the effect of removing the Coal- and Oil- specific broad fishing regions (e.g., section 112(c) list at any time that it Fired EGU source category from the Atlantic) to estimate the fraction of makes a negative appropriate and CAA section 112(c)(1) list because the methylmercury in commercially necessary finding under the section.’’ 70 EPA had not met the statutorily required CAA section 112(c)(9) delisting criteria. sourced fish caught in those broad FR 16032. In the 2005 Delisting Rule, Because coal- and oil-fired EGUs would regions attributable to U.S. EGUs. Both the EPA ‘‘identified errors in the prior remain on the CAA section 112(c)(1) of these simplifying assumptions mask [2000] finding and determined that the the potential complexity associated with source category list, the EPA proposed finding lacked foundation.’’ 70 FR to conclude that the CAA section 112(d) linking U.S. EGU-sourced mercury to 16032. Because the EPA concluded the methylmercury concentrations in these standards for that category, as 2000 Finding had been in error at the promulgated in the MATS rule, would commercial fish species. In particular, a time of listing, the Agency asserted that larger region such as the Atlantic likely be unaffected by the proposal if coal- and oil-fired EGUs ‘‘should never displays smaller-scale variation in finalized. have been listed under section 112(c) critical factors such as fish species In the proposal, the EPA requested and therefore the criteria of section habitat/location, patterns of mercury comment on two alternative 112(c)(9) do not apply’’ in removing the deposition, and factors related to the interpretations of the New Jersey source category from the list. Id. at methylation of mercury and associated holding. The first alternative 16033. Therefore, the EPA stated that it bioaccumulation/biomagnification. In interpretation probed whether the New developing these kinds of more had ‘‘inherent authority under the CAA Jersey decision does not apply because sophisticated models aimed at factoring to revise [the listing] at any time based the facts of the current situation are commercial fish consumption into a on either identified errors in the distinguishable from the underlying benefits analysis involving U.S. EGU December 2000 finding or on new facts of that case. Specifically, the EPA mercury, additional analyses could be information that bears upon that requested comment on the view that needed to understand this critical finding.’’ Id. at 16033. New Jersey would not apply because the element of spatial scale and the The D.C. Circuit rejected the EPA’s proposed reversal of the 2016 generalizing assumptions used by these interpretations, holding that the Agency Supplemental Finding is a continuation authors in linking mercury emissions did not have authority to remove source of the Agency’s response to the U.S. and deposition to commercial fish. Note categories from the CAA section 112(c) Supreme Court’s remand in Michigan. that in the EPA’s benefits analysis list based only on a revised CAA section Under this view, the Agency could rescind MATS without demonstrating completed for MATS, one reason focus 112(n)(1)(A) negative appropriate and that the CAA section 112(c)(9) criteria was placed on the freshwater angler necessary finding. The Court held that had been met because New Jersey did scenario was increased confidence in the CAA unambiguously requires the not address the situation in which the modeling this exposure pathway given EPA to demonstrate that the delisting Agency was revising its CAA section our ability to link patterns of U.S. EGU criteria in CAA section 112(c)(9) have 112(n)(1)(A) determination in response mercury deposition (relative to total been met before ‘‘any’’ source category to a U.S. Supreme Court decision. The deposition) over specific watersheds to can be removed from the CAA section second alternative interpretation sampled fish tissue concentrations in 112(c)(1) list. New Jersey, 517 F.3d at solicited comment on whether the EPA those same watersheds. This degree of 582. The D.C. Circuit specified that, would have the authority to rescind the refined spatial precision in linking U.S. under the plain text of the CAA, ‘‘the standards regulating HAP emissions EGU deposition to actual measured fish only way the EPA could remove EGUs under CAA section 112(d) in light of the tissue data increased overall confidence from the section 112(c)(1) list’’ was to fact that CAA section 112(n)(1)(A) in modeling benefits associated with satisfy those criteria. Id. The Court plainly requires that the Administrator this pathway, leading us to focus on the expressly rejected the EPA’s argument must find that regulation under CAA recreational angler exposure pathway. that, ‘‘[l]ogically, if EPA makes a section 112 is appropriate and necessary D. Effects of This Reversal of the determination under section as a prerequisite to undertaking such Supplemental Finding 112(n)(1)(A) that power plants should regulation. Under this theory, EGUs not be regulated at all under section 112 would remain on the CAA section 1. Summary of 2019 Proposal . . . [then] this determination ipso facto 112(c) list, but would not be subject to In the 2019 Proposal, the EPA must result in removal of power plants CAA section 112(d) standards, because proposed to conclude that finalizing a from the section 112(c) list.’’ Id. New Jersey did not address the question revision to the 2016 Supplemental (quoting the EPA’s brief). Instead, the of whether, in the absence of a valid and Finding to determine that it is not Court maintained that CAA section affirmative appropriate and necessary appropriate and necessary to regulate 112(n)(1) governed only how the finding, the EPA must regulate EGUs for HAP emissions from coal- and oil-fired Administrator determines whether to HAP. For both alternative EGUs would not lead to the removal of list EGUs, and that the EPA’s authority interpretations, the EPA solicited that source category from the CAA to remove a source category from the comment on whether the Agency had section 112(c)(1) list, nor would it affect list, even for EGUs, must be exercised the discretion to follow an alternative or the CAA section 112(d) standards only in accordance with the was, in fact, obligated to pursue an established in the MATS rule. requirements of CAA section 112(c)(9). alternative interpretation.

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00027 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 31312 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations

2. Final Rule valid appropriate and necessary meeting the delisting criteria of CAA After considering comments determination, which was the basis for section 112(c)(9). submitted in response to the EPA’s 2019 the EPA’s listing of the Coal- and Oil- Response: As explained in the 2019 Proposal, we are concluding that the Fired EGU source category. The Proposal, the EPA believes that the D.C. current action to reverse the 2016 commenters also argued that under the Circuit’s New Jersey decision governs Supplemental Finding would not affect plain meaning of the statutory text, the effects of the EPA’s final action. the CAA section 112(c) listing of EGUs Congress’ intention is clear that if the More specifically, this final action or the CAA section 112(d) regulations. EPA determines that regulation of EGU reversing the 2016 Supplemental The situation here is essentially emissions under CAA section 112 is not Finding does not remove the Coal- and indistinguishable to that in the New ‘‘appropriate and necessary,’’ then the Oil-Fired EGU source category from the Jersey case, and, therefore, in the EPA lacks jurisdiction to regulate such CAA section 112(c)(1) list. As the Court absence of the CAA section 112(c)(9) emissions. One commenter asserted that stated, ‘‘Congress . . . undoubtedly can delisting criteria being satisfied, coal- the EPA’s proposal to continue to limit an agency’s discretion to reverse and oil-fired EGUs necessarily remain enforce MATS while simultaneously itself, and in section 112(c)(9) Congress on the list of regulated sources, and the rejecting the factual and statutory basis did just that, unambiguously limiting CAA section 112(d) standards for the rule, offends the rule of law. EPA’s discretion to remove sources, promulgated in the MATS rule The commenters argued that the including EGUs, from the section necessarily remain in place. The EPA EPA’s reliance on the New Jersey 112(c)(1) list once they have been added did not propose a delisting analysis, and decision is misplaced because the to it.’’ 517 F.3d at 583. The Court the EPA does not intend to examine the regulatory landscape presented in this expressly rejected the argument made delisting criteria for the Coal- and Oil- action is fundamentally different than by the EPA at the time that if the Agency Fired EGU source category. Moreover, as what was assessed by the D.C. Circuit in reversed course and determined it was noted in the proposal, the results of the New Jersey. According to the not appropriate and necessary to CAA section 112(f)(2) residual risk commenters, the New Jersey decision regulate EGUs under CAA section 112, review conducted as part of this final only addressed the EPA’s authority to then that determination ‘‘logically’’ action indicate that with the MATS rule delist based on the reversal of an resulted in the removal of EGUs from in place, the estimated inhalation cancer appropriate and necessary finding the CAA section 112(c)(1) list. 517 F.3d risk to the individual most exposed to presumed to be legally valid, which is at 582. As the D.C. Circuit stated: ‘‘EPA’s disbelief that it would be actual emissions from the source a fact pattern not present in this action prevented from correcting its own category is 9-in-1-million, which would given the Michigan holding. One ‘errors’ except through section not satisfy the requirements for delisting commenter argued that because the EPA 112(c)(9)’s delisting process or court- as specified in CAA section 112(c)(9).57 had not yet issued any EGU HAP sanctioned vacatur cannot overcome the standards under CAA section 112(d) at 3. Comments and Responses plain text enacted by Congress.’’ 517 the time of New Jersey, the EPA’s F.3d at 583. Because coal- and oil-fired Comment: Some commenters argued interpretation of its regulatory EGUs remain on the CAA section that the EPA must rescind MATS if the jurisdiction under CAA section 112(n) 112(c)(1) source category list, the CAA Agency finalizes a determination that had not been subject to judicial review section 112(d) standards for the Coal- regulation under CAA section and the New Jersey decision, therefore, 112(n)(1)(A) is not appropriate and and Oil-Fired EGU source category, as does not speak to whether the EPA has promulgated in the MATS rule, are necessary. The commenters cited the authority to rescind a CAA section finding in Michigan which held that unaffected by this action. 112(d) standard after reversing the The EPA does not find persuasive ‘‘EPA interpreted [section 112(n)(1)(A)] appropriate and necessary finding. One unreasonably when it deemed cost commenters’ argument that New Jersey commenter further argued that to the is distinguishable because this action is irrelevant to the decision to regulate extent the EPA views its legal authority power plants’’ and asserted that if the not a reversal of a valid prior regarding continued enforcement of appropriate and necessary finding. As EPA now concludes that, based on a MATS to be ambiguous, it would be proper evaluation of costs, regulation of the commenters acknowledge, the D.C. arbitrary and capricious for the EPA to Circuit in New Jersey did not directly EGUs under CAA section 112 is not voluntarily leave MATS in place. appropriate and necessary, then either assess the validity of the EPA’s 2000 Conversely, there were many the CAA section 112(c) listing, the appropriate and necessary commenters who agreed with the EPA’s MATS rule, or both must be invalidated. determination. Rather, the EPA in its proposed approach to leave the MATS The commenters argued that, after the 2005 action revised the 2000 rule in place. These commenters agreed finalization of the proposal, there is no appropriate and necessary finding that the situation here is identical to because it was flawed. Similarly, here, 57 As relevant here, CAA section 112(c)(9) what was adjudicated in New Jersey; the EPA has determined that the 2016 provides that the ‘‘Administrator may delete any that is, in both cases (1) the EPA had Supplemental Finding was erroneous category from the list under this subsection . . . reversed an earlier final and effective (just as it did in 2005 with respect to the whenever the Administrator makes the following finding that regulating EGUs under CAA 2000 finding) and is finalizing reversal determination . . . (i) In the case of hazardous air pollutants emitted by sources in the category that section 112(n)(1)(A) was appropriate of the 2016 Supplemental Finding (just may result in cancer in humans, a determination and necessary, and (2) coal- and oil- as the EPA revised the 2000 finding). that no source in the category . . . emits such fired EGUs had been listed pursuant to We also disagree with the hazardous air pollutants in quantities which may CAA section 112(c). These commenters commenters’ argument that New Jersey cause a lifetime risk of cancer greater than one in one million to the individual in the population who concluded that following a final EPA is distinguishable because it was is most exposed to emissions of such pollutants determination that regulation of EGUs decided before the EPA had from the source . . . .’’ (emphases added). The under CAA section 112 is not promulgated a NESHAP for EGUs, and, findings of the EPA’s residual risk review indicate appropriate and necessary, both the therefore, the D.C. Circuit did not that it is extremely unlikely that any EPA Administrator could (much less would) lawfully CAA and the New Jersey holding are address the EPA’s authority to rescind exercise his or her discretion to ‘‘de-list’’ the Coal- clear that the only way to delist or de- MATS following a final determination and Oil-Fired EGU source category. regulate EGUs would be through that it is not appropriate and necessary

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00028 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations 31313

to regulate EGUs under CAA section include a full updating, subject to emit, or have the potential to emit, any 112. The statute does preclude a public comment, of the analytical data single HAP at a rate of 10 tons per year challenge to the EPA’s appropriate and base on which it rests,’’ EPN ‘‘formally (tpy) or more, or 25 tpy or more of any necessary finding until standards are in petitions EPA to continue the EGU combination of HAP. For major sources, place, see CAA section 112(e)(4); Util. MACT rule in effect’’ by making a new these standards are commonly referred Air Regulatory Grp. v. EPA, D.C. Cir. No. appropriate and necessary finding to as MACT (maximum achievable 01–1074, 2001 WL 936363 at *1 (D.C. ‘‘based on the facts as they stand today,’’ control technology) standards and must Cir., July 26, 2001), but nothing in the which EPN believes would support a reflect the maximum degree of emission D.C. Circuit’s reasoning in the New determination that regulation of EGUs reductions of HAP achievable after Jersey decision relied on the fact that the under CAA section 112 is appropriate considering cost, energy requirements, earlier appropriate and necessary and necessary. EPN comment at 36 and non-air quality health and finding was not yet reviewable. In New (April 17, 2019) (Docket ID Item No. environmental impacts. CAA section Jersey, the 2000 Finding was not yet EPA–HQ–OAR–2018–0794–2261). 112(d)(2) directs the EPA, in developing subject to judicial review and the EPA However, as explained above, the EPA MACT standards, to consider the argued that the inclusion of EGUs on the determines that this final action has no application of measures, processes, CAA section 112(c) list was not final effect on the MATS for EGUs; the MATS methods, systems, or techniques, Agency action; here, the 2016 rule remains in effect without any including, but not limited to, those that Supplemental Finding was final and further action by the EPA. To the extent reduce the volume of or eliminate HAP subject to judicial review. New Jersey is any response is needed, the EPA denies emissions through process changes, clear that, even following an EPA the EPN petition. substitution of materials, or other determination that it is not appropriate Comment: Numerous stakeholders modifications; enclose systems or and necessary to regulate EGUs under claimed a serious reliance interest in the processes to eliminate emissions; CAA section 112, the EPA cannot delist MATS rule that should weigh against collect, capture, or treat HAP when EGUs without going through the delisting or rescission of MATS as a released from a process, stack, storage, statutory delisting criteria (which the result of the EPA’s reversal of the 2016 or fugitive emissions point; are design, EPA has not done here). As long as Supplemental Finding. These equipment, work practice, or EGUs stay on the CAA section 112(c) stakeholders cited concerns about how operational standards; or any list of source categories, the EPA is delisting or rescission could lead to combination of the above. required to promulgate emission negative impacts on cost recovery of For these MACT standards, the statute standards under CAA section 112(d) significant capital investments, specifies certain minimum stringency regulating such sources. 42 U.S.C. potential disruptions to pre-existing air requirements, which are referred to as 7412(c)(2) (‘‘For the categories and quality planning efforts at the state- MACT floor requirements, and which subcategories the Administrator lists, level, or potentially foregone may not be based on cost the Administrator shall establish improvements in public health of the considerations. See CAA section emissions standards under subsection kind that have already resulted from 112(d)(3). For new sources, the MACT (d) of this section.’’). Thus, there is no improved air quality due to MATS floor cannot be less stringent than the question about it: Under the D.C. emissions reductions. Some emission control achieved in practice by Circuit’s holding in New Jersey, in order commenters pointed to these interests as the best-controlled similar source. The to rescind regulation under CAA section a reason why the EPA should not adopt MACT standards for existing sources 112(d), i.e., to rescind MATS, EGUs either of the two alternative can be less stringent than floors for new must first be delisted as a CAA section interpretations presented by the Agency sources, but they cannot be less 112(c) source category. in the 2019 Proposal regarding the stringent than the average emission As explained, the EPA believes that it potential effects of this Agency action. limitation achieved by the best- is bound by the D.C. Circuit’s New Response: The EPA’s revised performing 12 percent of existing Jersey decision. The New Jersey decision determination that regulation of EGUs sources in the category or subcategory itself was decided on Chevron step 1 under CAA section 112 is not (or the best-performing five sources for grounds. 517 F.3d at 582 (‘‘EPA’s appropriate and necessary will not categories or subcategories with fewer purported removal of EGUs from the remove EGUs from the CAA section than 30 sources). In developing MACT section 112(c)(1) list therefore violated 112(c) list of sources, and the previously standards, we must also consider the CAA’s plain text and must be established EGU MACT standard, as control options that are more stringent rejected under step one of Chevron.’’). established in MATS, remains in place. than the floor under CAA section Because the facts of this rulemaking are As a result, the EPA does not anticipate 112(d)(2). We may establish standards more stringent than the floor, based on substantially similar to those before the that any of the reliance interests cited D.C. Circuit in New Jersey, and because the consideration of the cost of above will be jeopardized as a result of the D.C. Circuit recognized that in such achieving the emissions reductions, any this action. a scenario the Agency has no discretion, non-air quality health and the EPA does not believe that it has any III. Background on the RTR Action environmental impacts, and energy discretion under Chevron, as one requirements. A. What is the statutory authority for commenter asserted, to voluntarily In the second stage of the regulatory this action? rescind MATS following this final process, the CAA requires the EPA to action. For these reasons, the EPA Section 112 of the CAA establishes a undertake two different analyses, which rejects commenters’ assertion that it is two-stage regulatory process to address we refer to as the technology review and acting in an arbitrary and capricious emissions of HAP from stationary the residual risk review. Under the manner in this determination of the sources. In the first stage, we must technology review, we must review the effect of this final Agency action. identify categories of sources emitting technology-based standards and revise The EPA additionally notes that one one or more of the HAP listed in CAA them ‘‘as necessary (taking into account commenter stated in its comment that if section 112(b) and then promulgate developments in practices, processes, the EPA finalized the proposal ‘‘based technology-based NESHAP for those and control technologies)’’ no less on any justification that does not sources. ‘‘Major sources’’ are those that frequently than every 8 years, pursuant

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00029 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 31314 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations

to CAA section 112(d)(6). Under the desulfurization systems and SO2 C. What changes did we propose for the residual risk review, we must evaluate continuous emissions monitoring Coal- and Oil-Fired EGU source the risk to public health remaining after systems installed and operational. category in our February 7, 2019, application of the technology-based Standards for filterable PM serve as a proposed rule? standards and must revise the surrogate for the non-mercury HAP On February 7, 2019, the EPA standards, if necessary, to provide an metals, with standards for total non- published a proposed rule in the ample margin of safety to protect public mercury HAP metals and individual Federal Register for the NESHAP for health or to prevent, taking into non-mercury HAP metals provided as Coal- and Oil-Fired EGUs, 40 CFR part consideration costs, energy, safety, and alternative equivalent standards. Work 63, subpart UUUUU, that took into other relevant factors, an adverse practice standards that require periodic consideration the RTR analyses. 84 FR environmental effect. The residual risk combustion process tune-ups limit 2670. In the proposed rule, we found review is required within 8 years after formation and emissions of the organic that residual risks due to emissions of promulgation of the technology-based HAP. standards, pursuant to CAA section air toxics from this source category are 112(f). In conducting the residual risk For oil-fired EGUs, the rule acceptable and that the current NESHAP review, if the EPA determines that the establishes standards to limit emissions provides an ample margin of safety to current standards provide an ample of HCl and hydrogen fluoride (HF), total protect public health, and we identified margin of safety to protect public health, HAP metals (e.g., mercury, nickel, lead), no new developments in HAP emission it is not necessary to revise the MACT and organic HAP (e.g., formaldehyde, controls to achieve further cost-effective standards pursuant to CAA section dioxin/furan). Standards for filterable emissions reductions under the 112(f).58 For more information on the PM serve as a surrogate for total HAP technology review. Based on the results statutory authority for this rule, see 84 metals, with standards for total HAP of these analyses, we proposed no FR 2670, February 7, 2019. metals and individual HAP metals revisions to the MATS rule. provided as alternative equivalent B. What is the Coal- and Oil-Fired EGU IV. What is included in this final rule standards. Periodic combustion process source category and how does the based on results of the RTR? tune-up work practice standards limit NESHAP regulate HAP emissions from formation and emissions of the organic This action finalizes the EPA’s the source category? determinations pursuant to the RTR HAP. The EPA promulgated the NESHAP provisions of CAA section 112 for the for Coal- and Oil-Fired EGUs The MATS rule was amended on Coal- and Oil-Fired EGU source (commonly referred to as MATS) on April 19, 2012 (77 FR 23399), to correct category. February 16, 2012 (77 FR 9304). The typographical errors and certain preamble text that was inconsistent with A. What are the final rule amendments standards are codified at 40 CFR part 63, based on the residual risk review for the subpart UUUUU. The MATS rule regulatory text; on April 24, 2013 (78 FR 24073), to update certain emission Coal- and Oil-Fired EGU source applies to existing and new coal- and category? oil-fired EGUs located at both major and limits and monitoring and testing area sources of HAP emissions. An EGU requirements applicable to new sources; We found risk due to emissions of air is a fossil fuel-fired combustion unit of on November 19, 2014 (79 FR 68777), to toxics to be acceptable from this source more than 25 megawatts (MW) that revise definitions for startup and category and determined that the serves a generator that produces shutdown and to finalize work practice current NESHAP provides an ample electricity for sale. A unit that standards and certain monitoring and margin of safety to protect public health cogenerates steam and electricity and testing requirements applicable during and prevent an adverse environmental supplies more than one-third of its periods of startup and shutdown; and effect. Therefore, we did not propose potential electric output capacity and on April 6, 2016 (81 FR 20172), to and are not finalizing any revisions to more than 25 MW electric output to any correct conflicts between preamble and the NESHAP for Coal- and Oil-Fired utility power distribution system for regulatory text and to clarify regulatory EGUs based on our analyses conducted sale is also an EGU. The source category text. In addition, the electronic under CAA section 112(f). covered by this MACT standard reporting requirements of the rule were B. What are the final rule amendments currently includes an estimated 713 amended on March 24, 2015 (80 FR based on the technology review for the EGUs located at approximately 323 15510), to allow for the electronic Coal- and Oil-Fired EGU source facilities. submission of Portable Document category? For coal-fired EGUs, the rule Format (PDF) versions of certain reports established standards to limit emissions until April 16, 2017, to allow for time We determined that there are no of mercury, acid gas HAP, non-mercury for the EPA’s Emissions Collection and developments in practices, processes, HAP metals (e.g., nickel, lead, Monitoring Plan System to be revised to and control technologies that warrant chromium), and organic HAP (e.g., accept all reporting that is required by revisions to the MACT standard for this formaldehyde, dioxin/furan). Standards the rule, and on April 6, 2017 (82 FR source category. Therefore, we did not for hydrochloric acid (HCl) serve as a 16736), and on July 2, 2018 (83 FR propose and are not finalizing revisions to the MACT standard under CAA surrogate for the acid gas HAP, with an 30879), to extend the interim section 112(d)(6). alternate standard for SO2 that may be submission of PDF versions of reports used as a surrogate for acid gas HAP for through June 30, 2018, and July 1, 2020, C. What are the effective and those coal-fired EGUs with flue gas respectively. compliance dates of the standards?

58 The D.C. Circuit has affirmed this approach to Additional detail regarding the The final rule is effective on May 22, implementing CAA section 112(f)(2)(A). See NRDC standards applicable to the seven 2020. No amendments to the MATS rule v. EPA, 529 F.3d 1077, 1083 (D.C. Cir. 2008) (‘‘If subcategories of EGUs regulated under are being promulgated in this action. EPA determines that the existing technology-based the MATS rule can be found in section Thus, there are no adjustments being standards provide an ’ample margin of safety,’ then the Agency is free to readopt those standards during IV.B of the 2019 Proposal. 84 FR 2670 made to the compliance dates of the the residual risk rulemaking.’’). (February 7, 2019). standards.

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00030 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations 31315

V. What is the rationale for our final Technology Review for the NESHAP for along with our proposed decisions decisions regarding the RTR action for Coal- and Oil-Fired EGUs Response to regarding risk acceptability, ample the Coal- and Oil-Fired EGU source Public Comments on February 7, 2019 margin of safety, and adverse category? Proposal, available in the docket for this environmental effects, in the February 7, This section of this preamble provides action. 2019, proposed rule. 84 FR 2697–2700. a description of what we proposed and A. Residual Risk Review for the Coal- The results of the risk assessment are what we are finalizing, the EPA’s and Oil-Fired EGU Source Category presented briefly in Table 2, and in rationale for the final decisions, and a more detail in the document titled summary of key comments and 1. What did we propose pursuant to Residual Risk Assessment for the Coal- responses. For comments not discussed CAA section 112(f) for the Coal- and and Oil-Fired EGU Source Category in in this preamble, comment summaries Oil-Fired EGU source category? Support of the 2019 Risk and and the EPA’s responses can be found Pursuant to CAA section 112(f), the Technology Review Proposed Rule (risk in the document titled Final EPA conducted a residual risk review document for the proposed rule), Supplemental Finding and Risk and and presented the results of this review, available in the docket for this action. TABLE 2—COAL- AND OIL-FIRED EGU INHALATION RISK ASSESSMENT RESULTS IN THE FEBRUARY 2019 PROPOSAL [84 FR 2670, February 7, 2019]

Maximum individual Population at Annual cancer Maximum chronic Maximum screening cancer risk increased risk of cancer incidence noncancer acute noncancer HQ 4 (in 1 million) 2 ≥1-in-1 million (cases per year) TOSHI 3 Number of Based on . . . Based on . . . Based on . . . Based on . . . facilities 1 Based on actual emission Actual Allowable Actual Allowable Actual Allowable Actual Allowable level emissions emissions emissions emissions emissions emissions emissions emissions level level level level level level level level

322 ...... 9 10 193,000 636,000 0.04 0.1 0.2 0.4 HQREL = 0.09 (arsenic). 1 Number of facilities evaluated in the risk analysis. There are an estimated 323 facilities in the Coal- and Oil-Fired EGU source category; however, one facility is lo- cated in Guam, which is beyond the geographic range of the model used to estimate risks. Therefore, the Guam facility was not modeled and the emissions for that facility are not included in this assessment. 2 Maximum individual excess lifetime cancer risk due to HAP emissions from the source category. 3 Maximum target organ-specific hazard index (TOSHI). The target organ systems with the highest TOSHI for the source category are neurological and reproduc- tive. 4 The maximum estimated acute exposure concentration was divided by available short-term threshold values to develop an array of hazard quotient (HQ) values. HQ values shown use the lowest available acute threshold value, which in most cases is the reference exposure level (REL). When an HQ exceeds 1, we also show the HQ using the next lowest available acute dose-response value.

a. Chronic Inhalation Risk Assessment emissions from oil-fired EGUs are the c. Multipathway Risk Screening and Results major contributor to the risk. The total Site-Specific Assessment Results estimated cancer incidence from this The results of the chronic inhalation source category, considering allowable Potential multipathway health risks cancer risk assessment based on actual emissions, is 0.1 excess cancer cases per under a fisher and gardener scenario emissions, as shown in Table 2 of this year, or one excess case in every 10 were identified using a three-tier preamble, indicate that the estimated screening assessment of the HAP known years. Based on allowable emissions, maximum individual lifetime cancer to be persistent and bio-accumulative in approximately 636,000 people are risk (cancer MIR) is 9-in-1 million, with the environment (PB–HAP) emitted by estimated to have cancer risks at or nickel emissions from oil-fired EGUs as facilities in the Coal- and Oil-Fired EGU the major contributor to the risk. The above 1-in-1 million from HAP emitted source category, and a site-specific total estimated cancer incidence from from the facilities in this source assessment of mercury using the EPA’s this source category is 0.04 excess category. The estimated maximum Total Risk Integrated Methodology.Fate, cancer cases per year, or one excess case chronic noncancer TOSHI for the source Transport, and Ecological Exposure in every 25 years. Approximately category is 0.4 (respiratory) based on (TRIM.FaTE) for one location (i.e., three 193,000 people are estimated to have allowable emissions, driven by facilities located in North Dakota) as cancer risks at or above 1-in-1 million emissions of nickel and cobalt from oil- further described below. Of the 322 from HAP emitted from the facilities in fired EGUs. No one is exposed to TOSHI MATS facilities modeled, 307 facilities this source category. The estimated levels above 1 based on allowable have reported emissions of carcinogenic maximum chronic noncancer TOSHI for emissions. PB–HAP (arsenic, dioxins, and the source category is 0.2 (respiratory), b. Screening Level Acute Risk polycyclic organic matter (POM)) that which is driven by emissions of nickel exceed a Tier 1 cancer screening value Assessment Results and cobalt from oil-fired EGUs. No one of 1, and 235 facilities have reported is exposed to TOSHI levels above 1 Table 2 of this preamble provides the emissions of non-carcinogenic PB–HAP based on actual emissions from sources worst-case acute HQ (based on the REL) (lead, mercury, and cadmium) that regulated under this source category. of 0.09, driven by emissions of arsenic. exceed a Tier 1 noncancer screening The EPA also evaluated the cancer There are no facilities that have acute value of 1. For facilities that exceeded risk at the maximum emissions allowed HQs (based on the REL or any other a Tier 1 multipathway screening value by the MACT standard (i.e., ‘‘allowable reference values) greater than 1. For of 1, we used additional facility site- emissions’’). As shown in Table 2 of this more detailed acute risk results, refer to specific information to perform an preamble, based on allowable assessment through Tiers 2 and 3, as the risk document for the proposed rule, emissions, the estimated cancer MIR is necessary, to determine the maximum available in the docket for this action. 10-in-1 million, and, as before, nickel chronic cancer and noncancer impacts

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00031 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 31316 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations

for the source category. For cancer, the d. Environmental Risk Screening modeling domain) did not exceed any highest Tier 2 screening value was 200. Results ecological benchmark. In addition, each This screening value was reduced to 50 An environmental risk screening individual modeled concentration of after the plume rise stage of Tier 3. assessment for the Coal- and Oil-Fired HCl and HF (i.e., each off-site data point Because this screening value was much EGU source category was conducted for in the modeling domain) was below the lower than 100-in-1 million, and the following pollutants: Arsenic, ecological benchmarks for all facilities. because we expect the actual risk to be cadmium, dioxins/furans, HCl, HF, lead, Based on the results of the lower than the screening value (site- mercury (methylmercury and mercuric environmental risk screening analysis, specific assessments typically lower chloride), and POMs. In the Tier 1 we do not expect an adverse estimates by an order of magnitude), we screening analysis for PB–HAP (other environmental effect as a result of HAP emissions from the Coal- and Oil-Fired did not perform further assessment for than lead, which was evaluated cancer. For noncancer, the highest Tier EGU source category. differently), POM emissions had no 2 screening value was 30 (for mercury), exceedances of any of the ecological e. Facility-Wide Risk Results with four facilities having screening benchmarks evaluated. Arsenic and values greater than 20. These screening An assessment of risk from facility- dioxin/furan emissions had Tier 1 values were reduced to 9 or lower after wide emissions was performed to exceedances for surface soil the plume rise stage of Tier 3. provide context for the source category Because the final stage of Tier 3 (time- benchmarks. Cadmium and risks. Based on facility-wide emissions series) was unlikely to reduce the methylmercury emissions had Tier 1 estimates developed using the same highest mercury screening values to 1, exceedances for surface soil and fish estimates of actual emissions for we conducted a site-specific benchmarks. Divalent mercury emissions sources in the source multipathway assessment of mercury emissions had Tier 1 exceedances for category, and emissions data from the emissions for this source category. sediment and surface soil benchmarks. 2014 National Emissions Inventory Analysis of the facilities with the A Tier 2 screening analysis was (NEI) (version 2) for the sources outside highest Tier 2 and Tier 3 screening performed for arsenic, cadmium, the source category, the estimated values helped identify the location for dioxins/furans, divalent mercury, and cancer MIR is 9-in-1 million, and nickel the site-specific assessment and the methylmercury emissions. In the Tier 2 emissions from oil-fired EGUs are the facilities to model with TRIM.FaTE. The screening analysis, arsenic, cadmium, major contributor to the risk. The total assessment took into account the effect and dioxin/furan emissions had no estimated cancer incidence based on that multiple facilities within the source exceedances of any of the ecological facility-wide emissions is 0.04 excess category may have on common lakes. benchmarks evaluated. Divalent cancer cases per year, or one excess case The three facilities selected are located mercury emissions from two facilities in every 25 years. Approximately near Underwood, North Dakota. All exceeded the Tier 2 screen for a 203,000 people are estimated to have three facilities had Tier 2 screening sediment threshold level benchmark by cancer risks at or above 1-in-1 million values greater than or equal to 20. Two a maximum screening value of 2. from HAP emitted from all sources at of the facilities are near each other (16 Methylmercury emissions from the the facilities in this source category. The kilometers (km) apart). The third facility same two facilities exceeded the Tier 2 estimated maximum chronic noncancer is more distant, about 20 to 30 km from screen for a fish (avian/piscivores) no- TOSHI posed by facility-wide emissions the other facilities, but it was included observed-adverse-effect-level (NOAEL) is 0.2 (respiratory), driven by emissions in the analysis because it is within the (merganser) benchmark by a maximum of nickel and cobalt from oil-fired EGUs. 50-km modeling domain of the other screening value of 2. A Tier 3 screening No one is exposed to TOSHI levels facilities and because it had an elevated assessment was performed to verify the above 1 based on facility-wide Tier 2 screening value. We expect that existence of the lake associated with emissions. These results are very similar the exposure scenarios we assessed for these screening values, and it was found to those based on actual emissions from these facilities are among the highest, if to be located on-site and is a man-made the source category because there is not not the highest, that might be industrial pond, and, therefore, was significant collocation of other sources encountered for other facilities in this removed from the assessment. with EGUs. Methylmercury emissions from two source category. The refined f. Proposed Decisions Regarding Risk multipathway assessment estimated an facilities exceeded the Tier 2 screen for a surface soil NOAEL for avian ground Acceptability, Ample Margin of Safety, HQ of 0.06 for mercury for the three and Adverse Environmental Effect facilities assessed. We believe the insectivores (woodcock) benchmark by a assessment represents the highest maximum screening value of 2. Other In determining whether risks are potential for mercury hazards through surface soil benchmarks for acceptable for this source category in fish consumption for the source methylmercury, such as the NOAEL for accordance with CAA section 112, the category. mammalian insectivores and the EPA considered all available health In evaluating the potential threshold level for the invertebrate information and risk estimation multipathway risk from emissions of community, were not exceeded. Given uncertainty. The risk results indicate lead compounds, rather than developing the low Tier 2 maximum screening that both the actual and allowable a screening threshold emission rate, we value of 2 for methylmercury, and the inhalation cancer risks to the individual compare maximum estimated chronic fact that only the most protective most exposed are well below 100-in-1 inhalation exposure concentrations to benchmark was exceeded, a Tier 3 million, which is the presumptive limit the level of the current NAAQS for lead environmental risk screen was not of acceptability. Also, the highest (0.15 micrograms per cubic meter). conducted for methylmercury. chronic noncancer TOSHI, and the Values below the level of the primary For lead, we did not estimate any highest acute noncancer HQ, are well (health-based) lead NAAQS are exceedances of the secondary lead below 1, indicating low likelihood of considered to have a low potential for NAAQS. For HCl and HF, the average adverse noncancer effects from multipathway risk. We did not estimate modeled concentration around each inhalation exposures. There are also low any exceedances of the lead NAAQS in facility (i.e., the average concentration risks associated with ingestion, with the this source category. of all off-site data points in the highest cancer risk being less than 50-

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00032 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations 31317

in-1 million based on a conservative 3. What key comments did we receive developing the RTR emissions dataset. screening assessment, and the highest on the residual risk review, and what For each organic HAP tested, if 40 noncancer hazard being less than 1 are our responses? percent or more of the available test data based on a site-specific multipathway The EPA received comments in were above test method detection limits, assessment. Considering this opposition to and in support of the emissions estimates for that HAP were information, the EPA proposed that the proposed residual risk review and our included in the modeling file. We assert residual risks of HAP emissions from determination that no revisions were that this approach which modeled each the Coal- and Oil-Fired EGU source warranted under CAA section 112(f)(2) organic HAP where up to 60 percent of category are acceptable. for the Coal- and Oil-Fired EGU source its 2010 ICR emissions data were below category. test method detection limits is a We then considered whether the reasonable and conservative means of current standards provide an ample Generally, the comments that were not supportive of the proposed estimating which organic HAP are margin of safety to protect public health determination from the risk review emitted from currently operating coal- and whether more stringent standards claimed that the risks are understated and oil-fired EGUs. We also assert that were necessary to prevent an adverse with the methods used by the EPA to the use of RDLs, which are based on environmental effect by taking into assess inhalation, multipathway, and averages of better-performing unit consideration costs, energy, safety, and environmental risks and suggested method detection levels, as well as other relevant factors. In determining changes to the underlying risk laboratories using the most sensitive whether the standards provide an ample assessment methodology. For example, analyses across many source categories, is a reasonable means of estimating margin of safety to protect public health, some commenters stated that the EPA organic HAP emissions from currently we examined the same risk factors that should lower the acceptability operating EGUs which, under the MATS we investigated for our acceptability benchmark so that risks below 100-in-1 rule, are not required to measure organic determination and also considered the million are unacceptable, include HAP emissions. With regard to the 2014 costs, technological feasibility, and emissions outside of the source category NEI organic HAP emissions estimates other relevant factors related to in question in the risk assessment, and referred to by commenters, the EPA emissions control options that might assume that pollutants with noncancer points out that those estimates are based health risks have no safe level of reduce risk associated with emissions on pre-MATS compliance information exposure. With regard to the Coal- and from the source category. In our and, thus, do not reflect reductions in analysis, we considered the results of Oil-Fired EGU source category risk organic HAP resulting from periodic the technology review, risk assessment, review, several commenters claimed tune-ups that have been conducted as and other aspects of our MACT rule that the type and quantity of organic required by the MATS rule. In addition, review to determine whether there are HAP emissions modeled were the pre-MATS compliance estimates in any cost-effective controls or other underestimated, disagreeing with the instances are likely to be based on, at measures that would reduce emissions EPA’s determination to model only 16 most, 19 site-specific tests which have further to provide an ample margin of organic HAP and to base the estimated an average ‘‘D’’ rating and which were safety. The risk analysis indicated that emissions on EPA-developed conducted over 25 years ago, as opposed the risks from the source category are representative detection levels (RDLs). to the MATS ICR data from up to 170 Commenters pointed to the difference low for both cancer and noncancer site-specific tests which would have an between the modeled 3.4 tons of total health effects, and, therefore, any risk average A rating and which were source category organic HAP emissions conducted just 9 years ago.59 Moreover, reductions from further available versus other estimates of total source control options would result in minimal the pre-MATS compliance estimates category organic HAP, such as the EPA’s most certainly includes emissions from health benefits. Moreover, no additional 2014 NEI estimate of over 3,000 tons of measures were identified for reducing EGUs that have since shut down. total source category organic HAP Although some comments were HAP emissions from affected sources in emissions from 130 organic HAP. supportive of the EPA’s proposed the Coal- and Oil-Fired EGU source The EPA disputes the comments determination based on results of the category. Thus, we proposed that the objecting to the type and quantity of risk review, the comments claimed that current MATS requirements provide an organic HAP modeled under the risk ample margin of safety to protect public review. As discussed in section IV.B of 59 As discussed in the Introduction to AP–42 (see health in accordance with CAA section the proposed rule (84 FR 2670, February https://www3.epa.gov/ttn/chief/ap42/c00s00.pdf), 112. 7, 2019), during the 2010 ICR effort for the AP–42 emission factor rating is an overall assessment of how good a factor is, based on both Based on the results of our the original MATS rulemaking process, the quality of the test(s) or information that is the environmental risk screening most of the organic HAP emissions data source of the factor and on how well the factor represents the emission source. A ‘D’ rated emission assessment, we also proposed that more for EGUs were at or below the detection levels of the prescribed test methods, factor is below average and is developed from test stringent standards are not necessary to data from a small number of facilities, and there even when long duration test runs (i.e., prevent an adverse environmental may be reason to suspect that these facilities do not approximately 8 hours) were required. represent a random sample of the industry. In effect. Under the MATS rule, organic HAP are addition, test data from ‘D’ rated emission factors may show evidence of variability within the source 2. How did the residual risk review regulated by a work practice standard population. Emission factors from the MATS ICR change for the Coal- and Oil-Fired EGU that requires periodic combustion have not been developed for AP–42 and the current source category? process tune-ups. As such, EGUs are not rating process has been revised from letter grades required to meet numeric emission to descriptors. However, under the previous rating Since proposal (84 FR 2670, February process, emission factors from the MATS ICR data limits for organic HAP or to test and would have received ‘A’ ratings, where an ‘A’ rated 7, 2019), neither the risk assessment nor report organic HAP emissions. Because emission factor is excellent and is developed from our determinations regarding risk the MATS rule does not require test data taken from many randomly chosen acceptability, ample margin of safety, or measurements of organic HAP, the EPA facilities in the industry population. Moreover, for an ‘A’ rated emissions factor, the source category adverse environmental effects have reviewed the available organic HAP test population is sufficiently specific to minimize changed. results from the 2010 ICR when variability.

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00033 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 31318 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations

the risks are overstated due to the overly 4. What is the rationale for our final adopted and forced widespread and conservative risk assessment approach and final decisions for the unprecedented coal-fired EGU methodology used by the EPA. residual risk review? retirements, that the general costs of Commenters stated, for example, that We evaluated all of the comments on emission control technologies have not the risk assessment makes numerous the EPA’s proposed residual risk review significantly been reduced and have conservative assumptions regarding and determined that no changes to the increased in some instances, and that emissions and exposures, the exposure review are needed. For the reasons the beyond-the-floor analyses conducted assumptions are scientifically outdated, explained in the proposed rule, we by the EPA in support of the 2012 and the assessment used unrealistically determined that the risks from the Coal- MATS Final Rule are still valid. high fish consumption rates. With and Oil-Fired EGU source category are Commenters also asserted that the EPA cannot adopt more stringent standards regard to the Coal- and Oil-Fired EGU acceptable, and that the current under CAA section 112(d)(6) where source category risk review, several standards provide an ample margin of safety to protect public health and there is no appreciable HAP-related commenters suggested data corrections benefit from doing so and pointed to the prevent an adverse environmental to emissions estimates for particular results of the risk assessment for the effect. Therefore, pursuant to CAA EGUs that, according to commenters, Coal- and Oil-Fired EGU source section 112(f)(2), we are finalizing our resulted in overstated emissions being category. modeled. One commenter also residual risk review as proposed. The comments that were not suggested several revisions to the B. Technology Review for the Coal- and supportive of the proposed emissions estimation methodology for Oil-Fired EGU Source Category determination from the technology HAP emissions from EGUs. Several review generally claimed that the 1. What did we propose pursuant to commenters pointed out that the EPA’s review failed to assess whether control CAA section 112(d)(6) for the Coal- and risk review for the Coal- and Oil-Fired technologies deployed for compliance Oil-Fired EGU source category? EGU source category and the June 2018 with the 2012 MATS Final Rule were Electric Power Research Institute (EPRI) Pursuant to CAA section 112(d)(6), less expensive and more effective than risk studies for coal-fired power the EPA conducted a technology review, projected and whether technologies plants 60—each of which followed which focused on identifying and deemed economically infeasible in 2012 somewhat different methodologies— evaluating developments in practices, have since become cheaper. The EPA disagrees with the comments similarly concluded that human health processes, and control technologies for the emission sources in the source opposing the proposed determination risks associated with HAP emissions are category. After conducting the CAA that no revisions were warranted under within EPA acceptability thresholds. section 112(d)(6) technology review of CAA section 112(d)(6). As explained in The EPA acknowledges that the risk the NESHAP for Coal- and Oil-Fired section VI.C of the proposed rule (84 FR assessment results for the Coal- and Oil- EGUs, we proposed that revisions to the 2670, February 7, 2019), control Fired EGU source category are standards are not necessary because we technologies typically used to minimize dependent on the emission values used identified no cost-effective emissions of pollutants that have in the assessment. If we were to lower developments in practices, processes, or numeric emission limits under the emission rates based on more accurate control technologies. More information MATS rule include electrostatic data, we expect lower risk estimates. concerning our technology review is in precipitators and fabric filters for Because the EPA has determined that the memorandum titled Technology control of PM and non-mercury HAP the risk is acceptable, and that the Review for the Coal- and Oil-Fired EGU metals; wet scrubbers and dry scrubbers existing standards provide an ample Source Category, available in the docket for control of acid gases (SO2, HCl, and margin of safety to protect public health for this action, and in the February 7, HF); and activated carbon injection for control of mercury. These existing air in accordance with CAA section 112, 2019, proposed rule. 84 FR 2700. pollution control technologies that are making the data corrections suggested 2. How did the technology review currently in use are well-established by commenters would potentially change for the Coal- and Oil-Fired EGU and provide the capture efficiencies reduce risk further but would not source category? necessary for compliance with the change the determinations under the Since proposal (84 FR 2670, February MATS emission limits. Organic HAP, risk review. Accordingly, we conclude 7, 2019), the technology review has not including emissions of dioxins and that it is reasonable not to update the changed. furans, are regulated by a work practice risk assessment following the proposal, standard that requires periodic burner and we have finalized the risk 3. What key comments did we receive tune-ups to ensure good combustion. document and re-submitted it to the on the technology review, and what are This work practice continues to be a docket for this action as the Residual our responses? practical approach to ensuring that Risk Assessment for the Coal- and Oil- The EPA received comments in combustion equipment is maintained Fired EGU Source Category in Support support of and against the proposed and optimized to run to reduce of the 2019 Risk and Technology Review technology review and our formation and emissions of organic HAP Final Rule. determination that no revisions were and continues to be expected to be more warranted under CAA section 112(d)(6) effective than establishing a numeric 60 EPRI. June 8, 2018. Hazardous Air Pollutants for the Coal- and Oil-Fired EGU source standard for emissions that, due to (HAPs) Emission Estimates and Inhalation Human category. current detection levels, cannot reliably Health Risk Assessment for U.S. Coal-Fired Electric The comments that agreed with the be measured or continuously monitored. Generating Units: 2017 Base Year Post-MATS EPA’s proposed determination that no We received no comments that included Evaluation. Available at https://www.epri.com/#/ revisions to the MATS rule are specific information on costs or pages/product/3002013577/?lang=en.EPRI. June 22, warranted based on results of the performance for control technologies 2018. Multi-Pathway Human Health Risk Assessment for Coal-Fired Power Plants. Available technology review also asserted that the deployed to comply with the 2012 at https://www.epri.com/#/pages/product/ reductions required by MATS were not MATS Final Rule or for other control 3002013523/?lang=en. cost-effective at the time they were technology, work practices, operational

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00034 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations 31319

procedures, process changes, or F. What analysis of environmental G. What analysis of children’s pollution prevention approaches that justice did we conduct? environmental health did we conduct? reduce HAP emissions. Since proposal, Executive Order 12898 (59 FR 7629, This action is not subject to Executive no information has been presented to Order 13045 because it is not February 16, 1994) establishes federal cause us to change the proposed economically significant as defined in executive policy on environmental determination that no developments in Executive Order 12866, and because the justice. Its main provision directs practices, processes, or control EPA does not believe the environmental technologies, nor any new technologies federal agencies, to the greatest extent health or safety risks addressed by this or practices were identified for the practicable and permitted by law, to action present a disproportionate risk to control of non-mercury HAP metals, make environmental justice part of their children. This action’s health and risk acid gas HAP, or mercury, and that no mission by identifying and addressing, assessments are summarized in section developments in work practices nor any as appropriate, disproportionately high V.A of this preamble and are further new work practices or operational and adverse human health or documented in sections V and VI of the procedures have been identified for the environmental effects of their programs, proposed rule (84 FR 2670, February 7, control of organic HAP. policies, and activities on minority 2019), and the risk document for the populations and low-income final rule,62 available in the docket for 4. What is the rationale for our final populations in the U.S. this action. approach for the technology review? As discussed in section VI.A of the VII. Statutory and Executive Order We evaluated all of the comments on proposed rule (84 FR 2670, February 7, Reviews 2019), to examine the potential for any the EPA’s technology review and Additional information about these determined that no changes to the environmental justice issues that might be associated with the source category, statutes and Executive Orders can be review are needed. For the reasons found at https://www.epa.gov/laws- we performed a demographic analysis, explained in the proposed rule, we regulations/laws-and-executive-orders. determined that no cost-effective which is an assessment of risk to developments in practices, processes, or individual demographic groups of the A. Executive Orders 12866: Regulatory control technologies were identified in populations living within 5 km and Planning and Review and Executive our technology review to warrant within 50 km of the facilities.61 In the Order 13563: Improving Regulation and revisions to the standards. Therefore, analysis, we evaluated the distribution Regulatory Review pursuant to CAA section 112(d)(6), we of HAP-related cancer and noncancer This action is a significant regulatory are finalizing our technology review as risks from the Coal- and Oil-Fired EGU action that was submitted to OMB for proposed. source category across different review because it is likely to raise novel demographic groups within the legal or policy issues. Any changes VI. Summary of Cost, Environmental, populations living near facilities. The made in response to OMB and Economic Impacts and Additional results of the Coal- and Oil-Fired EGU recommendations have been Analyses Conducted source category demographic analysis documented in the docket. The EPA A. What are the affected facilities? indicate that emissions from the source does not project any potential costs or category expose approximately 193,000 benefits associated with this action. The EPA estimates that there are 713 people to a cancer risk at or above 1-in- B. Executive Order 13771: Reducing existing coal- and oil-fired EGUs located 1 million and no people to a chronic Regulations and Controlling Regulatory at 323 facilities that are subject to the noncancer TOSHI greater than 1. There Costs MATS rule and will be affected by this are only four facilities in the source final action. category with cancer risk at or above 1- This action is not considered an in-1 million, and all of them are located Executive Order 13771 regulatory B. What are the air quality impacts? in Puerto Rico. Consequently, all of the action. There are no quantified cost percentages of the at-risk population in estimates for this final rule because it Because the EPA is not promulgating will not result in any changes in costs. any amendments to the MATS rule, each demographic group associated there will be no air quality impacts as with the Puerto Rican population are C. Paperwork Reduction Act (PRA) a result of this final action. much higher than their respective This action does not impose any new nationwide percentages, and those not information collection burden under the C. What are the cost impacts? associated with Puerto Rico are much PRA. OMB has previously approved the lower than their respective nationwide Because the EPA is not promulgating information collection activities percentages. The methodology and the any amendments to the MATS rule, contained in the existing regulations results of the demographic analysis are there will be no cost impacts as a result and has assigned OMB control number presented in the technical report titled of this final action. 2060–0567. This action does not impose Risk and Technology Review—Analysis an information collection burden D. What are the economic impacts? of Demographic Factors for Populations because the EPA is not making any Living Near Coal- and Oil-Fired EGUs changes to the information collection Because the EPA is not promulgating Regulated Under the Mercury and Air requirements. any amendments to the MATS rule, Toxics Standards (MATS), available in there will be no economic impacts as a Docket ID No. EPA–HQ–OAR–2018– D. Regulatory Flexibility Act (RFA) result of this final action. 0794. I certify that this action will not have E. What are the benefits? a significant economic impact on a 61 See technical report titled Risk and Technology Because the EPA is not promulgating Review—Analysis of Demographic Factors for 62 See document titled Residual Risk Assessment any amendments to the MATS rule, Populations Living Near Coal- and Oil-Fired EGUs for the Coal- and Oil-Fired EGU Source Category in Regulated Under the Mercury and Air Toxics Support of the 2019 Risk and Technology Review there will be no benefits as a result of Standards (MATS). May 23, 2018; Docket ID Item Final Rule, available in Docket ID No. EPA–HQ– this final action. No. EPA–HQ–OAR–2018–0794–0012. OAR–2018–0794.

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00035 Fmt 4701 Sfmt 4700 E:\FR\FM\22MYR2.SGM 22MYR2 31320 Federal Register / Vol. 85, No. 100 / Friday, May 22, 2020 / Rules and Regulations

substantial number of small entities The Blue Lake Rancheria stated that the document for the final rule, available in under the RFA. This action will not EPA’s cost-benefit analysis should not the docket for this action (see document impose any requirements on small exclude co-benefits, and that the titled Residual Risk Assessment for the entities. The EPA does not project any analysis should include healthcare costs Coal- and Oil-Fired EGU Source potential costs or benefits associated and environmental remediation costs. Category in Support of the 2019 Risk with this action. The tribe discussed the health effects of and Technology Review Final Rule, exposure to mercury and noted that the E. Unfunded Mandates Reform Act available in Docket ID No. EPA–HQ– RTR shows that the risks are acceptable (UMRA) OAR–2018–0794). with MATS in place; that margin of This action does not contain any safety would be eliminated if the rule is I. Executive Order 13211: Actions unfunded mandate as described in rescinded. The tribe also expressed Concerning Regulations That UMRA, 2 U.S.C. 1531–1538, and does concern that eliminating the MATS rule Significantly Affect Energy Supply, not significantly or uniquely affect small will prolong the use of coal-fired power Distribution, or Use governments. The action imposes no plants, which would lead to increased enforceable duty on any state, local, or This action is not a ‘‘significant greenhouse gas emissions. energy action’’ because it is not likely to tribal governments or the private sector. The EPA held a consultation with the have a significant adverse effect on the Fond du Lac Band of Lake Superior F. Executive Order 13132: Federalism supply, distribution, or use of energy. Chippewa on April 3, 2019. The tribe This action is not anticipated to have This action does not have federalism also did not support the proposed impacts on energy supply decisions for implications. It will not have substantial finding that regulation of HAP the affected electric utility industry. direct effects on the states, on the emissions from coal- and oil-fired EGUs relationship between the national is not appropriate and necessary. The J. National Technology Transfer and government and the states, or on the tribe stated that studies have shown that Advancement Act (NTTAA) distribution of power and mercury is harmful and should be responsibilities among the various controlled, and that the EPA does not This action does not involve technical levels of government. have the authority to delist EGUs from standards. G. Executive Order 13175: Consultation regulation under CAA section 112. K. Executive Order 12898: Federal and Coordination With Indian Tribal According to the tribe, co-benefits from Actions To Address Environmental Governments reductions of non-HAP pollutants Justice in Minority Populations and should be considered equally with Low-Income Populations This action does not have tribal benefits from reductions of HAP. The implications as specified in Executive tribe asked whether the EPA had The EPA believes that this action does Order 13175. It would neither impose considered factors specific to their tribe substantial direct compliance costs on not have disproportionately high and in the EPA’s analysis, such as their adverse human health or environmental tribal governments, nor preempt Tribal higher consumption of fish due to law. Thus, Executive Order 13175 does effects on minority populations, low- cultural and subsistence reasons and the income populations, and/or indigenous not apply to this action. prevalence of wetlands and ditches on Consistent with the EPA Policy on peoples, as specified in Executive Order the reservation, which are mercury Consultation and Coordination with 12898 (59 FR 7629, February 16, 1994). sinks. The tribe also questioned whether Indian Tribes, the EPA consulted with The documentation for this decision is impacts to wildlife such as otters, loons, tribal officials during the development contained in section VI.F of this and eagles were considered. of this action. A summary of the preamble, section VI.A of the proposed Responses to these comments and rule (84 FR 2670, February 7, 2019), and consultations follows. others received are available in the RTC On April 2, 2019, the EPA held a the technical report, Risk and document,63 available in the docket for consultation with the Blue Lake Technology Review—Analysis of this action. Rancheria. The tribe indicated that they Demographic Factors for Populations did not support the 2019 Proposal for H. Executive Order 13045: Protection of Living Near Coal- and Oil-Fired EGUs several reasons. The tribe expressed Children From Environmental Health Regulated Under the Mercury and Air concern that the EPA’s proposed finding Risks and Safety Risks Toxics Standards (MATS), available in that it is not appropriate and necessary This action is not subject to Executive the docket for this action (see Docket ID to regulate HAP emissions from coal- Order 13045 because it is not Item No. EPA–HQ–OAR–2018–0794– and oil-fired EGUs under section 112 of economically significant as defined in 0012). the CAA would remove the legal Executive Order 12866, and because the L. Congressional Review Act (CRA) foundation for the MATS rule. The tribe EPA does not believe the environmental added that the EPA has neither the health or safety risks addressed by this This action is subject to the CRA, and authority nor the obligation to remove action present a disproportionate risk to the EPA will submit a rule report to coal- and oil-fired EGUs from the CAA children. This action’s health and risk each House of the Congress and to the section 112(c) source category list or to assessments are contained in sections V Comptroller General of the United rescind MATS. The tribe noted that the and VI of the proposed rule (84 FR 2670, States. This action is not a ‘‘major rule’’ costs of compliance for EGUs subject to February 7, 2019), and the risk as defined by 5 U.S.C. 804(2). MATS have already been incurred, and Dated: April 16, 2020. that those investments could be in vain 63 See document titled Final Supplemental if MATS is rescinded. In addition, the Finding and Risk and Technology Review for the Andrew Wheeler, proposed finding will likely lead to NESHAP for Coal- and Oil-Fired EGUs Response to Administrator. Public Comments on February 7, 2019 Proposal, litigation which would be a waste of available in Docket ID No. EPA–HQ–OAR–2018– [FR Doc. 2020–08607 Filed 5–21–20; 8:45 am] taxpayer dollars, according to the tribe. 0794. BILLING CODE 6560–50–P

VerDate Sep<11>2014 18:24 May 21, 2020 Jkt 250001 PO 00000 Frm 00036 Fmt 4701 Sfmt 9990 E:\FR\FM\22MYR2.SGM 22MYR2