Document omkbDV9dMjMq42220mE2MR0XD

Morgan Lewis Ronald J. Tenpas [ ] PartneErx!"6 Ronrirnp"s@ m organlew is.com BY ELECTRONIC FILING http:// www.regulations.gov Docket No. EPA-HQ-SFUND-2015-0781 July 11, 2017 Barbara Foster and Michael Pease Program Implementation and Information Division Office of Resource Conservation and Recovery Mail Code 5303P U.S. Environmental Protection Agency 1200 Pennsylvania Ave., NW Washington, DC 20460 Re: Comments on Proposed Rule, Financial Responsibility Requirements under CERCLA 108(b) for Classes of Facilities in the Hard rock Mining Industry, 82 Fed. Reg. 3388 (Jan. 11, 2017) Dear Ms. Foster and Mr. Pease: I. Introduction These comments are submitted on behalf of the American Chemistry Council, American Fuel & Petrochemical Manufacturers, American Iron & Steel Institute, Cement Kiln Recycling Coalition, Industrial Minerals Association - North America, National Association of Manufacturers, National Mining Association, North American Metals Council, Portland Cement Association, Superfund Settlements Project, The Fertilizer Institute, and U.S. Chamber of Commerce. This coalition formed due to common concerns related to the Proposed Rule, Financial Responsibility Requirements under CERCLA 108(b) for Classes of Facilities in the Hardrock Mining Industry, 82 Fed. Reg. 3388 (Jan. 11, 2017) (the "Proposed Rule"). Collectively these organizations represent more than 3 million businesses spanning all sizes, sectors and regions of the country. Having reviewed the Proposed Rule and EPA's associated justification and explanation, the coalition members share very deep concerns with EPA's proposal and regulatory approach. The flaws in EPA's approach, discussed in more detail below, are of concern not only to those parties engaged in the Sierra Club v. EPA 18cv3472 NDCA Morgan, Lewis & Bockius LLP mi Pennsylvania Ave., NW Washington, DC 20004-2541 United States O +1.202.739.3000 +1.202.739.3001 Tier 7 ED 002061 00086796-00001 Docket No. EPA-HQ-SFUND-2015-0781 July 11, 2017 Page 2 hardrock mining and mineral processing ("HRM") sector but also to those engaged in other industrial sectors because the flaws are illustrative of the additional failures that will follow if EPA stands by its current proposal and then applies the approach to other sectors down the road. Most notably, EPA's approach fails to ground the regulatory assessment in a proper and detailed analysis of modern risk conditions associated with contemporary operating practices and, thus, greatly overstates potential risks and liabilities in the HRM industry. At the same time, EPA essentially ignores the current and comprehensive regulatory regime to which the HRM industry is subject, a regime that addresses operations, spill response, closure and reclamation, and financial responsibility requirements. In doing so, the Proposed Rule seeks to address the same risks the states and federal regimes already address. These fundamental flaws in EPA's approach have particular relevance to this coalition given that all of our industries are already subject to strict and diverse federal and state operating and financial responsibility requirements ranging from environmental review under the National Environmental Policy Act ("NEPA") or similar state environmental review laws, to chemical manufacture registration and use requirements, to required spill prevention and containment programs and inspections, to mandatory reporting of chemical releases, to Risk Management and Process Safety Management regulations, to closure, corrective action and financial assurance requirements for operating facilities under the Resource Conservation and Recovery Act ("RCRA"), to federal and state-based reclamation and closure programs and associated financial assurance. Given these existing and multi-layered regulatory requirements, the Proposed Rule seeks, and any similar rules for other sectors would be seeking, to address contingencies that will not occur and certainly have not been demonstrated to regularly recur. Section 108(b) of the Comprehensive Environmental Response, Compensation, and Liability Act ("CERCLA"), 42 U.S.C. 9608(b), permits EPA to adopt or decline to adopt rules requiring that certain "classes of facilities establish and maintain evidence of financial responsibility," provided also that any such rules must not be more than what is required to be "consistent with the degree and duration of risk associated with the production, transportation, treatment, storage, or disposal of hazardous substances." 42 U.S.C. 9608(b)(1) (emphasis added). Yet, EPA's approach in the Proposed Rule results in a financial responsibility requirement that is unnecessary and not calibrated to the "degree and duration of risk" posed. Accordingly, the coalition urges EPA to withdraw the Proposed Rule, recognizing that there is not sufficient risk to warrant any new CERCLA financial responsibility regulations. It is not surprising that EPA's proposal and analysis should suffer from such defects given that the Proposed Rule resulted from a rushed process with very little stakeholder input. From the regulated community, to states, to the financial industry and to the Small Business Administration, not a single entity was adequately engaged in the development of the Proposed Rule. Similarly, the very limited peer review that occurred was conducted without transparency and so late in the process EPA could not practically have addressed the myriad issues the peer review raised. As a consequence, EPA has proposed a rule riddled with errors and based on insufficient and unrepresentative data. Sierra Club v. EPA 18cv3472 NDCA Tier 7 ED 002061 00086796-00002 Docket No. EPA-HQ-SFUND-2015-0781 July 11, 2017 Page 3 Given the many fundamental defects relating to the Proposed Rule's substance and development, the Proposed Rule must be withdrawn as it is arbitrary, capricious, an abuse of discretion, and otherwise not in accordance with law and is in excess of statutory jurisdiction, authority and limitation and short of statutory right. See Administrative Procedure Act, 5 U.S.C. 500 et seq. ("APA"). As shown in these comments, and those submitted by many others, a proper evaluation of the degree and duration of risk, accompanied by appropriate attention to the many concerns raised about the Proposed Rule, should lead EPA to conclude that no CERCLA 108(b) financial responsibility rule is required. II. EPA's determination of continuing risk is indefensible. EPA's statutory directive allows it to adopt rules requiring certain "classes of facilities establish and maintain evidence of financial responsibility consistent with the degree and duration of risk associated with the production, transportation, treatment, storage, or disposal of hazardous substances." 42 U.S.C. 9608(b)(1) (emphasis added). The statute further states, "[t]he level of financial responsibility shall be initially established, and, when necessary, adjusted to protect against the level of risk which the President in his discretion believes is appropriate based on the payment experience of the Fund, commercial insurers, courts settlements and judgments, and voluntary claims satisfaction . . . " 42 U.S.C. 9608(b)(2). The plain language of the statute is very clear: whether to impose financial responsibility requirements at all is discretionary but, if that discretion is exercised, any rule must be limited to addressing "the degree and duration of risk" and based on "the payment experience of the Fund, commercial insurers, courts settlements and judgments, and voluntary claims satisfaction."1 The Proposed Rule fails to satisfy the statutory criteria. a. EPA ignores CERCLA's mandate to assess the "degree and duration of risk." The statute's specific direction is embodied in the "degree and duration" limitation. Congress made clear that it wanted any 108(b) regulation to be justified by the real risk of such events. See, e.g., https://www.epa.gov/superfund/frequent-questions-about-financial-responsibilitv-requirements-undercomprehensive: Johnson Declaration at HH 6,16. Thus, the statute directs EPA to conduct a risk analysis, 1That EPA previously identified hardrock mining as a class of facilities for which requirements would be first developed does not diminish the Administrator's discretion regarding whether to impose financial responsibility requirements and, if requirements are imposed, which facilities to exclude from such requirements. See 82 Fed. Reg. 3398; Opposition of Respondent United States Environmental Protection Agency to Petitioners' Petition for Writ of Mandamus, Appendix A, Declaration of Barnes Johnson at HH26-29, 39, In re Idaho Conservation League, No. 14-1149 (DC Cir. Nov. 19, 2014) (hereinafter, "Johnson Declaration"); see also Identification of Priority Classes of Facilities for Development of CERCLA 108(b) Financial Responsibility Requirements, 74 Fed. Reg. 37,213 (July 28, 2009) (the "Priority Notice"). As set forth in the notice, additional research and outreach to stakeholders were necessary before publication of a proposed rule. Johnson Declaration at H26. EPA cannot "use a notice issued before even a proposed rule, to make any final determinations." Johnson Declaration at 1128. EPA can only finalize a regulation if "necessary" based on risk. Johnson Declaration at 1|29. Indeed, EPA has already invoked its discretion, altering the universe of potentially regulated facilities from those identified in the initial Priority Notice. See, e.g., 82 Fed. Reg. 3400, 3454-59. Sierra Club v. EPA 18cv3472 NDCA Tier 7 ED 002061 00086796-00003 Docket No. EPA-HQ-SFUND-2015-0781 July 11, 2017 Page 4 and because such risk is obviously related to the operating practices and regulatory regimes currently in place, the statute clearly requires EPA to assess current operating practices and existing regulatory requirements regarding those operations (including permitting, operating, release containment and response, closure, and financial assurance regulations, among others). Only with such a comprehensive evaluation can EPA determine if risk exists and, if so, the degree of that risk, including whether any release is likely to leave an unmet financial obligation to address any clean-up activity. EPA has lost sight of what the statute directs and, instead, has crafted a Proposed Rule that seeks to provide assurance against "all potential risk." 82 Fed. Reg. 3405. For example, EPA makes the unreasonable assumption that every site feature at every operating facility will require remediation under CERCLA.2 82 Fed. Reg. 3405. See U.S. EPA, Response to Peer Review Comments: CERCLA 108(b) Financial Responsibility Formula for Hardrock Mining Facilities Background Document, Draft, December 2016 ("Response to Peer Review Comments") at 3-5 ("the formula conservatively assumes that all portions of a facility could require a CERCLA response tomorrow" while admitting even at historic National Priorities List sites "costs were incurred only at the portion of the facility resulting in a CERCLA response"). EPA makes this assumption despite its failure to produce even a single site where, in fact, each and every site feature required remediation. This assumption is disconnected from reality and renders the Proposed Rule arbitrary and capricious. EPA also improperly relies on irrelevant and historical information to establish risk and overstates the risk those sites have actually demonstrated. At the same time, EPA ignores that there are existing, pervasive regulatory programs that (1) require owners and operators to adopt safe operating practices to avoid releases, (2) implement programs and designs to minimize the scope of releases when they do occur, and (3) address any resulting environmental conditions the day they occur, at closure, and during post-closure. Many of these requirements are backed by financial assurance, while others must be addressed as needed and a failure to do so could result in fines and penalties. In ignoring these current-day features of modern day operations, EPA has failed to conduct the kind of analysis the statute requires. b. EPA's use of inappropriate current and historic data exaggerates the risks of a release and any likely costs of remediation. i. EPA exaggerates the risk of release by relying on data that are not instructive. EPA gathered data from the Emergency Response Notification System ("ERNS"), the Toxic Release Inventory ("TRI"), and the Resource Conservation and Recovery Act Biennial Reports ("RCRA BR") to establish the presence of hazardous substances at mining sites in a misleading and distorted attempt 2 EPA collected information on response costs from 319 sites and costs of specific activities at 438 operable units at 88 sites to generate an estimate of total response costs at the facilities. It then linked specific site features to releases or threatened releases and to remedies. EPA ultimately identified thirteen site features that served as potential sources of release that resulted in remedies within twelve categories. 82 Fed. Reg. 3462. EPA's formula estimates cost for remediation of every site feature to be funded through financial responsibility instruments. Sierra Club v. EPA 18cv3472 NDCA Tier 7 ED 002061 00086796-00004 Docket No. EPA-HQ-SFUND-2015-0781 July 11, 2017 Page 5 to demonstrate risk. 82 Fed. Reg. 3476-78. Mere presence does not equal risk. EPA took a similarly flawed approach in seeking to justify its 2009 Priority Notice and 2010 Identification of Additional Classes of Facilities for Development of Financial Responsibility Requirements under CERCLA Section 108(b), 75 Fed. Reg. 817 (Jan. 6, 2010) (the "Identification Notice"). In the Proposed Rule and each of its earlier notices, EPA relied on the data as proof of risk where those data prove nothing of the sort. The ERNS collects only initial accounts of releases reported to the National Response Center and, thus, those reports are preliminary and of varying reliability. In addition, the statutes and regulations requiring reporting have low reporting thresholds, so the existence of reporting cannot support any risk conclusions related to a likely future unfunded clean-up obligation. Further, this dataset does not include any analysis of the response actions taken by the facilities, meaning it is divorced from any relevant risk evaluation. While the releases reported may be common and unavoidable for many industries, see 82 Fed. Reg. 3477, EPA gives no meaningful acknowledgement of the fact that companies are also legally required and equipped to respond to them at the time they occur, including being subject to follow up inspections by state and federal regulators. Finally, and importantly, these reports show that, with only thirty releases per year since 2000, the HRM industry is well-regulated and responsive. See 82 Fed. Reg. 3477. The TRI includes estimates of the volume of hazardous substances that the industry released, recycled, transferred, treated or used for energy recovery. Importantly, the volumes reported include permitted releases and disposal at, or discharge to, permitted facilities, as well as significant recycling activities. In the case of the HRM industry, 85 to 95% of the reported volumes are trace amounts of naturally occurring metal and metal compounds that are present in rock and dirt at mine facilities and that are managed in engineered facilities permitted and regulated by federal and state law. TRI estimates were similarly considered in the Priority Notice and the Identification Notice, a use that industry objected to based on EPA's own caution regarding the misuse of TRI data. Specifically, EPA has repeatedly noted that the TRI release estimates alone are not sufficient to determine exposure levels or to calculate potential risks to human health and the environment. See, e.g., https://www.epa.gov/toxics-release-inventorv-tri-orogram/what-can-tri-tell-vou-about-risk; see also U.S. EPA, TRI National Analysis 2015 (Updated January 2017), available at https://www.epa.gov/sites/production/files/2017-01/documents/tri na 2015 complete english.pdf: U.S. EPA, "The Toxics Release Inventory (TRI) and Factors to Consider When Using TRI Data" at 4), available at https://www.epa.gov/toxics-release-inventorv-tri-program/factors-consider-when-using-toxics-release-i nventorv-data (setting forth five key factors to consider when using TRI data, including that toxicity varies among the covered chemicals; the presence in the environment must be evaluated with the potential and actual exposures and routes of exposures; many options for managing production-related wastes are subject to stringent technical standards and exacting regulatory oversight; regulatory controls apply to many of the releases; and some wastes may be double-counted because of reporting by both the generator and handler). Notwithstanding all of this, in the Proposed Rule, EPA asserts that the presence of hazardous substances provides "some" indication of the "potential" for risk "if' improperly managed. Sierra Club v. EPA 18cv3472 NDCA Tier 7 ED 002061 00086796-00005 Docket No. EPA-HQ-SFUND-2015-0781 July 11, 2017 Page 6 82 Fed. Reg. 3477. Given EPA's own caveats on how limited the TRI data are in demonstrating any actual risk, citing TRI data does not satisfy the robust "degree and duration" analysis that the statute requires.3 EPA takes a similarly failed path with reference to RCRA BR data. These data simply show quantities of hazardous wastes generated and managed in accordance with law. The generation of hazardous waste does not correlate to risk of an actual or threatened release of a hazardous substance that requires a CERCLA remedy. Again, this complete lack of connection with risk was pointed out to EPA in response to the Identification Notice, yet EPA continues to maintain that presence alone can be instructive on actual risk. Finally, EPA attempts to bolster its "continuing risk" determination through a series of reports on releases of hazardous substances at HRM sites. These EPA reports are significantly flawed and undermine, rather than support, this determination. As detailed in the National Mining Association's comments on the Proposed Rule, these sites fall into three major categories: (1) sites that are legacy sites and are simply not comparable to currently operating mining facilities; (2) sites where a release did occur but where it is being promptly remediated by the owner or operator without burdening the taxpayer; and (3) sites where a release occurred and state or federal programs have responded and made regulatory changes to address related future conditions. In sum, the information EPA has relied upon to justify its conclusion that the HRM industry presents a "continuing risk" and therefore must be subject to 108(b) requirements is not indicative of the reality stemming from current operations, nor could that data serve to demonstrate real risk for any other industry. ii. EPA fails to consider the risk reducing effects of modern regulatory regimes, the duplication it will create, and the preemptive impact its program could have on existing state regimes. When 108(b) was enacted in 1980, there were few environmental laws, and those that existed were in their infancy. As EPA stated, "past operating procedures, before the advent of environmental laws, were likely in many cases to give rise to environmental problems that current regulations and modern operating practices can prevent or minimize." 82 Fed. Reg. 3461 (emphasis added). By 2017, the historic regulatory void has been filled with comprehensive and robust federal and state regulatory regimes, including regimes that have evolved and improved since 108(b)'s initial adoption by Congress. 3Separate and apart from the general unsuitability of the TRI data as a surrogate for risk, EPA has erroneously relied on the TRI data from "Iron and Steel Mills" (NAICS code 331110) as a surrogate for hazardous pollutant releases to the environment from the iron ore mining industry (NAICS code 212210). The TRI reports from Iron and Steel Mills identified every U.S. steel mill (including integrated, electric arc furnace, coke, finishing mills, etc.) as falling within the iron ore mining category, and inaccurately estimated 52 million pounds of chemical releases, including numerous chemicals that are simply not associated with the iron mining process. Had EPA actually assessed TRI reporting for the iron ore mining sector, it would have found that iron ore mining was actually excluded by EPA after the industry demonstrated that releases were below reporting thresholds. Sierra Club v. EPA 18cv3472 NDCA Tier 7 ED 002061 00086796-00006 Docket No. EPA-HQ-SFUND-2015-0781 July 11, 2017 Page 7 The HRM industry, like the other sectors EPA may seek to regulate in the future, is now subject to environmental review under NEPA, 42 U.S.C. 4321 etseq., and similar state environmental review laws; media-specific programs addressing hazardous substances, such as through programs established under the Federal Water Pollution Control Act, 33 U.S.C. 1251 etseq., Clean Air Act, 42 U.S.C. 7401 etseq., and the Solid Waste Disposal Act, 42 U.S.C. 6901 etseq.; manufacturing-specific programs, such as chemical registration and review under the Toxic Substances Control Act, 15 U.S.C. 2601 et seq.; operating-specific programs, such as Risk Management Plan and Process Safety Management regulations and storage tank regulations; and general emergency response programs, such as under both CERCLA and the Emergency Planning and Community Right-to-Know Act, 42 U.S.C. 11001 etseq. While the HRM industry uses hazardous substances and occasionally has releases, that does not mean that the existing regulatory scheme is not effective. The record indicates that the existing programs are working to address the limited risk currently posed and that every feature at every mine does not fail. To the extent there may be releases, nothing suggests that the HRM industry presents a level of risk that fits within CERCLA's "highest level of risk of injury" and thus fits within CERCLA's direction on where to target regulation. EPA can certainly exercise its discretion to conclude that isolated risk scenarios that might yet be imagined do not warrant or require the whole sector being burdened with an expensive and unnecessary financial responsibility regime. HRM operations must undertake an extensive permit process to obtain approvals from the appropriate agencies-the United States Bureau of Land Management ("BLM"), the United States Forest Service ("USFS") and the state. These permits address all aspects of operation, including mining, beneficiation, mineral processing, reclamation, closure and post-closure care, and contain strict financial assurance requirements. Permits under these programs cannot be secured if the applicant cannot demonstrate that it will comply with numerous design and operational requirements to minimize the risk of significant spills or other releases that could adversely impact the environment. For example, all western states in which mining occurs have staff dedicated to ensuring that mining facilities are designed, constructed, and operated to minimize risk to the environment and ensure reclamation and closure are completed. See Letter from Matthew H. Mead, Governor of Wyoming and Chairman, Western Governors' Association, and Steve Bullock, Governor of Montana and Vice Chair, Western Governors' Association, to Honorable Gina McCarthy, Administrator, United States Environmental Protection Agency, March 29, 2016 ("March WGA Comments"). There is evidence that these programs work. The Small Business Administration Office of Advocacy reports that the BLM and USFS both report no National Priorities List ("NPL") listings for the thousands of mines either agency has approved since 1990, and that states have had to call only a few bonds for small sums. See, Letter from The Honorable Darryl L. DePriest, Chief Counsel, Office of Advocacy, U.S. Small Business Administration, to The Honorable Gina McCarthy, Administrator, U.S. Environmental Protection Agency, Re: Financial Responsibility Requirements for the Hardrock Mining Industry (Docket ID: EPA-HQ-SFUND-2015-0781) at 4, January 19, 2017 (EPA-HQ-SFUND-2015-0781-1406) ("Advocacy Comments"). Sierra Club v. EPA 18cv3472 NDCA Tier 7 ED 002061 00086796-00007 Docket No. EPA-HQ-SFUND-2015-0781 July 11, 2017 Page 8 These state and federal programs are especially significant given Congress' evident intent that EPA not over-regulate or duplicate. Congress expressed this intent in multiple ways. For example, CERCLA directs EPA to impose requirements "for facilities in addition to those under Subtitle C of [RCRA] and other Federal law." 42 U.S.C. 9608(b)(l). Similarly, Congress directed EPA to develop a plan to avoid requiring financial assurances that are duplicative of those already required by other federal agencies in its 2014 EPA appropriations bill, H.R. 2279,113th Cong. 103-105 (engrossed in H.R. Jan. 9, 2014). 160 Cong. Rec. H475, H979 (daily ed. Jan. 15, 2014). While EPA maintains that the Proposed Rule is "in addition to" and "effectively complement^]" these other programs, 82 Fed. Reg. 3402, EPA does not support this claim by identifying where the other programs fall short. Nor could it because these programs address risk to such a degree that any CERCLA 108(b) program is unnecessary. EPA concedes that a "highly-developed regulatory landscape [] is already in place for hard rock mining," Johnson Declaration at H55, and "federal closure programs [] may have an effect on the risk a facility presents." U.S. EPA, CERCLA 108(b) Hardrock Mining and Mineral Processing Evaluation of Markets for Financial Responsibility Instruments, and the Relationship of CERCLA 108(b) to Financial Responsibility Programs of Other Federal Agencies ("MCS") at 3, August 25, 2016. Further, EPA "recognizes that, in requiring implementation of controls pursuant to [the Federal program] objectives, some federal mine closure program requirements help to address releases to the environment and thereby may have the effect of reducing the risk a facility presents." MCS at 6. There is no "may" about it - if a program addresses releases to the environment, which CERCLA seeks to address, that program "will" have the effect of reducing risk. The same can be said for state programs. CERCLA contains an express preemption provision that demonstrated Congress' long-standing concern to avoid duplication of state or local laws: Except as provided in this subchapter, no owner or operator of a vessel or facility who establishes and maintains evidence of financial responsibility in accordance with this subchapter shall be required under any State or local law, rule, or regulation to establish or maintain any other evidence of financial responsibility in connection with liability for the release of a hazardous substance from such vessel or facility. Evidence of compliance with the financial responsibility requirements of this subchapter shall be accepted by a State in lieu of any other requirement of financial responsibility imposed by such State in connection with liability for the release of a hazardous substance from such vessel or facility. 42 U.S.C. 9614. EPA has long recognized that CERCLA 108(b) rules would have federalism implications. See Johnson Declaration at 1135. Recognizing that many states have financial responsibility requirements, EPA sought to "carefully consider the effects of its CERCLA section 108(b) rules on other programs to Sierra Club v. EPA 18cv3472 NDCA Tier 7 ED 002061 00086796-00008 Docket No. EPA-HQ-SFUND-2015-0781 July 11, 2017 Page 9 avoid any unanticipated consequences." Johnson Declaration at f]13. State programs are well-established and comprehensive, now having decades of experience and having developed an "expertise" with mining regulation, compared to EPA's minimal experience in this regulatory space. See 82 Fed. Reg. 3401; Western Governors' Association, Policy Resolution 2014-07 at 2. Nevertheless, EPA chose to ignore both Congress' clear intent to avoid duplication and the robust programs already in place that deal with the same risks EPA is seeking to address under the Proposed Rule. Instead, it has proposed a duplicative, one size fits all financial regulatory program that is overbroad, and which will cause conflict, create unnecessary expense and spur litigation. This is not simply a concern of the commercial enterprises that will be impacted - it is also a matter of importance to the state governments whose programs are being replicated. See March WGA Comments (additional requirements would impair western economies, the industry and fail to address local requirements necessary to protect the environment, particularly water resources). For example, a significant overlap exists between the site features addressed under the federal and state hard rock mining programs and site features that form the basis for the formula in the Proposed Rule. The engineering cost estimates in EPA's financial responsibility formula are based on the exact activities conducted as part of closure and reclamation under existing parallel state and federal regulations. In fact, EPA's Response Data Collection used a sample of 63 facilities' reclamation and closure plan engineering cost data to develop its formula. 82 Fed. Reg. 3463; RIA at 3-8. Furthermore, EPA acknowledged, "some closure programs conduct activities that reduce CERCLA risks." 82 Fed. Reg. 3401. Peer Review Commenter 4 also noted the duplication between what EPA is assuming will be future remedial costs and the activities that companies are already required to undertake under current regulatory regimes: In the end I have no idea what the relevance is of any of the data presented in Section 2.2 [CERCLA Response Activities]. On page 2-15 EPA states "EPA's prior experience with CERCLA cleanups leads it to expect that similar types of remedies will continue to be selected for mining facilities in the future." There is no reason to make any presumptions here - the closure and reclamation plans and data collected in Section 3 indicate exactly what types of remedies are required at current HMFs. The engineering studies relied upon categorize the expense categories (tailings, leach dumps, pit, hazard removal, indirect costs, direct costs, etc., pp. 2-19 - 2-20). Perhaps the idea is that EPA was looking for justification for its methodology in Section 3, feeling it needed to prove that relying on company engineering plans was reasonable and that companies would not be leaving anything important out. Sierra Club v. EPA 18cv3472 NDCA Tier 7 ED 002061 00086796-00009 Docket No. EPA-HQ-SFUND-2015-0781 July 11, 2017 Page 10 Response to Comments at 6-5. EPA even acknowledges that there is definite duplication between the Proposed Rule and existing state and federal regulations by holding out the prospect of potential reductions in financial assurance obligations at the back end to account for the existing programs.4 But that is backwards under the statute EPA has an obligation to study and recognize the effectiveness of these programs before fashioning regulations that will presumptively apply, not impose an overbroad regulatory regime on the promise that there is a case by case "off ramp" that will solve any duplication its new rules create.5 It is EPA's obligation, in the first instance, to demonstrate that its proposed program is appropriately tailored to and consistent with the statute. This includes an obligation to discuss and assess in detail the other regulatory regimes, which it has failed to do. Those regimes are substantial and, once EPA acknowledges this, it will be clear that EPA can forego any further requirements. EPA is not required to eliminate "all potential risks" and should not require facilities to secure financial responsibility instruments to cover "all potential risks." CERCLA does not suggest or require such an expansive reading of "risk." EPA has the authority and responsibility to draw the line between acceptable and unacceptable risks and then only require financial responsibility for those unacceptable risks according to their "degree and duration." In fact, EPA may lawfully determine that the risks from certain classes of facilities, for whatever reasonable bases, do not warrant financial responsibility requirements at all. Finally, EPA asserts in a single paragraph that it makes "sound policy sense" to read the statute and the Proposed Rule as not pre-empting state programs. This is nonsense. If the proposed program duplicates the state requirements, then Section 114(d) mandates that an overlapping state program be pre-empted. In simply hypothesizing that no conflict will exist, EPA has failed to address, and has essentially chosen to ignore, one other aspect of the problem - the likelihood that its Proposed Rule will have preemptive effect, with all the attendant policy and legal problems this generates in a federal system that is designed to insure a primary role for the states. State and federal regimes have evolved substantially over the last several decades, creating a much different federalism and pre-emption posture than might have been the case had EPA undertaken this rulemaking at a much earlier date. But EPA did wait, and those state regimes have emerged. EPA's failure to develop a detailed assessment of those state regimes necessarily means EPA has arbitrarily and capriciously failed to do a serious preemption analysis related to its proposal, given that such a base-line state regime assessment is essential to any preemption analysis. That failure makes the Proposed Rule infirm. 4As discussed later in these comments, however, the proposed reduction methodology is so poorly constructed that any reductions do not appear to be available in reality. 5 ironically, EPA rejected a site-specific approach for the assessment of risks because it was too "resource intensive to implement," 82 Fed. Reg. at 3460, but has simultaneously claimed that the same site-specific approach is appropriate for evaluation for release from the 108(b) program because the agency "has substantial experience making individualized determinations of site risk, as this practice is consistent with EPA's practice under the Superfund program." 82 Fed. Reg. at 3415. Sierra Club v. EPA 18cv3472 NDCA Tier 7 ED 002061 00086796-00010 Docket No. EPA-HQ-SFUND-2015-0781 July 11, 2017 Page 11 iii. EPA's reliance on historic data overstates potential frequency and severity of releases from current facilities. In the Proposed Rule and supporting documents, EPA regularly relies on data from sites that are the product of operations that occurred decades ago to estimate the future scope and clean-up cost from currently operating facilities. In large part, EPA claims that this reliance is appropriate given the similarities between current and historic mining practices. Specifically, EPA concludes that "although some mining waste practices have changed over time, the basic technology for extraction and beneficiation of mineral ores have remained fairly constant over the last 50 years." 82 Fed. Reg. 3475. From this oversimplification, EPA alleges that releases today will be the same in scope and impact as those that occurred historically. EPA wholly ignores the evolution of mining practices and state and federal oversight over the last 30 years, including new technologies, controls and monitoring to minimize, prevent and sometimes eliminate the risk the Proposed Rule ostensibly seeks to address.6 EPA's overly simplistic conclusion about the similarities in mining practices also contradicts other statements in the proposal regarding the effectiveness of current regulations. As noted in the prior section, EPA recognizes that "past operating procedures, before the advent of environmental laws, were likely in many cases to give rise to environmental problems that current regulations and modern operating practices can prevent or minimize." 82 Fed. Reg. 3461 (emphasis added). There is abundant evidence regarding the extensive nature of regulations governing today's hardrock mining industry at both the state and federal levels. See Western Governors' Association Policy Resolution 2014-07 and Policy Resolution 2014-09, attached to and incorporated by reference in Letter from Steve Bullock, Governor of Montana and Chair, Western Governors' Association, and Dennis Daugaard, Governor of South Dakota and Vice Chair, Western Governors' Association, to Honorable Gina McCarthy, Administrator, United States Environmental Protection Agency, August 17, 2016 ("August WGA Comments"); Advocacy Comments at 4-7. The evolution of the legal and regulatory environment, and industry technological innovations, have completely changed the landscape and undercut the Agency's conclusions that superficial similarities in mining practices mean that the risks ofthe past persist. Given the significant advancements in mining practices, in the event of a release requiring remediation, such remediation would be expected to have a lesser risk and burden profile than that 6 EPA also incorrectly assumes significant CERCLA remediation liabilities for low-risk facilities, e.g., iron ore mining, even though there have been no iron ore mining sites on EPA's NPL; there have been no federal government expenditures related to CERCLA remediation at these sites; iron ore mines use largely inert materials; and all ofthe facilities are small quantity generators of hazardous waste. Further, EPA has excluded the Iron ore mining industry from reporting requirements under the Emergency Planning and Community Right-To-Know Act (EPCRA) Section 313 Toxic Release Inventory (TRI) because EPA concluded: 1) the extraction and beneficiation of iron ore do not routinely use hazardous substances to produce a final product, and toxic chemical releases and transfers are not of sufficient quantities to warrant reporting; 2) no facilities were expected to meet the threshold reporting levels under EPCRA; and 3) iron ore mining and associated facilities do not make extensive use of toxic chemicals for processing their product. Additional discussion on low-risk facilities can be found in the American Iron and Steel Institute's comment letter on the draft rule. Sierra Club v. EPA 18cv3472 NDCA Tier 7 ED 002061 00086796-00011 Docket No. EPA-HQ-SFUND-2015-0781 July 11, 2017 Page 12 from much older operations.7 Indeed, two of the peer reviewers of EPA's formula raised concerns about the Agency's use of historic data to justify conclusions on the costs of remediation. Peer Review Commenter 1 to EPA's proposal stated, "[o]ne concern with using Actual [historic] Response Cost data from the NPL is that compared to data from currently operating facilities, the facilities on the NPL might be more costly than currently operating facilities to remediate (e.g., you have to be above a threshold in the hazard ranking system). So using NPL sites would result in an overestimate of costs." Response to Peer Review Comments at 3-1. Similarly, Peer Review Commenter 4 concluded: In the end I have no idea what the relevance is of any of the data presented in Section 2.2 [CERCLA Response Activities]. On page 2-15 EPA states "EPA's prior experience with CERCLA cleanups leads it to expect that similar types of remedies will continue to be selected for mining facilities in the future." There is no reason to make any presumptions here-the closure and reclamation plans and data collected in Section 3 indicate exactly what types of remedies are required at current HMFs. EPA must heed such criticisms. III. EPA conducts a flawed assessment of benefits. EPA identified the following "benefits" of the Proposed Rule: "a reduction in costs the government must bear to fulfill cleanup obligations, improved environmental practices at mining sites, avoided impacts to impaired waters, and faster cleanups." 82 Fed. Reg. 3396. EPA readily admits that the reduction in government costs is the only measurable benefit. 82 Fed. Reg. 3395-96; RIA at ES-12. The discussion of the other benefits is peppered with words like "may" and "potential." At best, these benefits are speculative and, upon close review, they are completely absent. EPA's claim of reduction in government cost is flawed in multiple respects. First, as discussed below, the cost estimate relies on an assumption that the response costs estimated are actually representative of costs associated with future CERCLA cleanups. They are not. 82 Fed. Reg. 3394. Moreover, EPA used a high-end estimate that assumes exiting firms fail to meet any of their clean-up obligations. Id. EPA assumed that all the firms that failed would, in every instance, leave behind CERCLA liabilities and aII thirteen site features will require a response action and absolutely no funds would be available to address any of them. Id. There is no support for such a grossly overstated assumption. EPA also failed to quantify what the cost of the rule will be to the government. In particular, EPA did not quantify government administrative costs, see RIA at ES-7, 3-6,4-11, although these costs could be substantial between development of the electronic portal EPA proposes, on-going review of confidential business information requests, maintenance of the public facing website, and review of financial responsibility calculations (including requests for reductions and release) and procured 7 Even with such historic sites, not every site feature generated a CERCLA response action, further undermining the formula in the Proposed Rule that works on a contrary supposition. Sierra Club v. EPA 18cv3472 NDCA Tier 7 ED 002061 00086796-00012 Docket No. EPA-HQ-SFUND-2015-0781 July 11, 2017 Page 13 financial responsibility instruments. In short, EPA has proposed a regulatory program yet ignored the fundamental costs associated with its implementation. With regard to EPA's claim that the Proposed Rule would provide an incentive for companies to invest in and implement additional environmental protection technologies, the reality is that such incentives already exist due to the various state and federal regulatory regimes discussed above. Instead, the Proposed Rule would divert funds from other economically beneficial uses, including investments in evolving technology and pollution prevention, by requiring that large amounts of capital assets be put aside to protect against a hypothetical risk in the future, risk that is already covered by the current regulatory regime. Additionally, the reduction criteria are so fatally flawed that they may perversely induce facilities to adopt controls not even relevant to their day-to-day operations. It is not just capital investment that may be impacted by the diversion of funds for financial responsibility requirements. EPA failed to consider the impacts on jobs and the economy by tying up such large amounts of capital. Ironically, in its justification for providing standard wording of financial instruments, EPA states it "does not wish to create a situation where resources that otherwise would have been devoted to cleanups would be expended," see 82 Fed. Reg. 3416; yet that is what this proposal, or any proposal like it, will do. EPA's third potential benefit piggybacks on its claim that the Proposed Rule will result in improved environmental practices: "To the extent that the Proposed Rule leads to improvements in facilities' environmental performance, the rule may reduce acid mine drainage and other discharges into waterways." (Emphasis added). 82 Fed. Reg. 3396. EPA admits that any such benefit is "contingent upon changes in behavior among regulated entities to reduce the environmental risk." 82 Fed. Reg. 3396. Accordingly, the same reasoning that applied to the prior "benefit" applies here - the current regulatory scheme is protective, including addressing any releases to waterways, and the diversion of significant funds is likely to result in productive investment being stifled. EPA's last claimed benefit is also illusory. The suggestion that this Proposed Rule will, in the future, produce faster clean-ups in meaningful numbers relies on the unfounded assumptions that: (1) a large number of solvent firms will default; and (2) that a cleanup will even be necessary. EPA's failure to conduct a proper risk analysis to determine the likelihood of these events in a modern regulatory regime clouds its identification of perceived benefits. Moreover, the slow pace of Superfund cleanups has been notorious - no matter the funding mechanism - belying any suggestion that changing financial responsibility requirements will improve clean-up times. See, e.g., Memorandum, from E. Scott Pruitt to Deputy Administrator, General Counsel, Assistant Administrators, Inspector General, Chief Financial Officer, Chief of Staff, Associate Administrators, Regional Administrators, re: Prioritizing the Superfund Program, May 22, 2017. IV. EPA's approach to analyzing costs is arbitrary and capricious, dramatically underestimates costs to industry and fails to effectively analyze other impacts. Sierra Club v. EPA 18cv3472 NDCA Tier 7 ED 002061 00086796-00013 Docket No. EPA-HQ-SFUND-2015-0781 July 11, 2017 Page 14 The lack of environmental benefit is particularly glaring when considering the estimated cost to implement the Proposed Rule. EPA is approaching the height of arbitrary and capricious agency action in proposing a rule that it acknowledges to be cost inefficient. Moreover, the estimated costs appear to be unduly low as a consequence of EPA relying on inadequate data, using a formula laden with errors and unsupportable assumptions, and then compounding all these errors by extrapolating to the full universe from an unrepresentative modeled universe. a. EPA's estimated costs of compliance are staggeringly high and subject to serious question. EPA estimates the annual compliance cost to industry to be $171 million, without allowing the use of the financial test, or $111 million, with the financial test.8,9 82 Fed. Reg. 3393. These costs do not even include the significant cost that would be incurred in obtaining the financial responsibility instruments. By omitting these expenses, EPA greatly underestimates the total cost and annual costs of procuring instruments under the Proposed Rule. In contrast, EPA estimates that the potential cost to the Government would be a maximum of $527 million over 34 years, or $15.5 million per year. 82 Fed. Reg. 3395. To compare: Annual Cost Cost over 34 years Cost to Industry $111 million to $171 million $7.1 billion Government Cost Avoided $15.5 million $527 million This cost/benefit disparity is staggering but is even more disturbing because EPA's cost estimates are substantially underestimated.8910 The estimates EPA has produced are based on insufficient data to accurately model both economic and societal costs. EPA estimates that 221 facilities, owned by 121 ultimate parent companies, would be subject to the Proposed Rule (the "full universe"), 8 That the estimated costs are so high is hardly surprising when considering the extreme assumption discussed above that every company will default and every site feature will require remediation. 9These amounts also do not appear to include the cost to maintain the company website envisioned by the proposed rule. See 82 Fed. Reg. 3393 (listing the administrative costs to industry). These requirements are extensive, intrusive and will result in significant administrative costs, particularly with regard to claims of confidential business information ("CBI"). Indeed, the level of public notice in the proposed rule is excessive. Based on CBI considerations and issues related to competitive advantage, wholesale publication on a company or EPA website of all information the owner or operator is required to submit to EPA, whether CBI or not, would be burdensome and damaging to businesses. For example, publication of this information may invite lawsuits and, in addition, may confuse the public because the sites subject to the Proposed Rule are not current Superfund sites. Similarly, providing for public notice and comment for EPA determinations to lower the amount of financial responsibility, or to fully release financial responsibility, is unnecessary. Such decisions do not appear to be the type of decisions requiring such a complicated, time-consuming and costly public comment procedure, the costs of which do not appear to be accounted for in EPA's calculations. 10As just one example, EPA's calculated financial responsibility liability for the entire universe of regulated facilities (221) was $7.1B, but when the correct values and acreages are used in EPA's financial responsibility calculation, the resulting financial responsibility for the eight active iron mining operations alone appears to approach approximately $8.5B. Sierra Club v. EPA 18cv3472 NDCA Tier 7 ED 002061 00086796-00014 Docket No. EPA-HQ-SFUND-2015-0781 July 11, 2017 Page 15 although it acknowledges that it may not have accurately identified all facilities that would be subject to the Proposed Rule. 82 Fed. Reg. 3390. Moreover, EPA only obtained facility-specific data for 49 facilities (22%) and financial information for only 21 companies (17%), all of which are publicly traded firms and own only 38 of the facilities (17%) (the "modeled universe"). 82 Fed. Reg. 3390-92. EPA further assumes that the full universe would have "similar financial characteristics" of the modeled universe, but the preamble acknowledges that there are "uncertainties" with that supposition. 82 Fed. Reg. 3393. The Regulatory impact Analysis provides a clear picture of just how substantial these uncertainties are. The mining facilities that would be subject to the rule "have a diverse group of owners, including sole proprietors, domestic and multinational corporations, and joint-venture partnerships" with many privately held. RIA at 2-8. The firms also varied substantially in size approximately 31% of firms each having fewer than 100 employees while the five largest firms have more than 70,000 employees. RIA at 2-9. For the subset of companies for which data was available, 25% of firms each had less than $15 million in annual revenues, while 25% of firms each had more than $4 billion in annual revenues. RIA at 2-9. EPA thus concedes that the financial data that were available "may not be representative of smaller entities in the regulated universe" because information was more readily available for large, publicly-traded companies. RIA at 2-9. The Small Business Administration Office of Advocacy indicates that 36% of hardrock mining businesses are small businesses. Advocacy Comments at 2. b. The financial responsibility formula is indefensible. EPA's overly simplistic formula bears no relation to reality. EPA focused on just three variables (area, precipitation and flow) ultimately using only just one (area) to determine the cost for most site features, as if any industrial site can be reduced to such simplistic terms. EPA relied upon data that were inappropriate, with much of it obsolete and/or unrepresentative. EPA only had a small data set, which is particularly troubling where the data show widely varying costs and/or include a significant outlier. EPA inappropriately excluded certain data, biasing high its estimates. EPA incorporated unexplained assumptions, such as its use of smear factors. EPA used "hypothetical" cost information that "should not be considered a perfect substitute for actual financial assurance pricing." RIA at 4-4. There are also errors within the formula. As detailed in the Office of Advocacy's comments, the results from the formula are unrealistic. See Advocacy Comments at 9-10. Peer Review Commenter 4 could not replicate a single cost number when conducting spot checks, had "grave concerns about the integrity of the data," and "suspected there are errors that result from a misunderstanding of the information in the original source documents stemming from a lack of consultation with industry." Response to Peer Review Comments at 2-8 - 2-9. Sierra Club v. EPA 18cv3472 NDCA Tier 7 ED_002061_00086796-00015 11, Docket No. EPA-HQ-SFUND-2015-0781 July 2017 Page 16 Importantly, the proposed financial responsibility reduction methodology, which EPA claims will incentivize environmental improvements and reduce the required level of financial responsibility, is based on a review of a limited number of programs and a poor understanding and misapplication of existing reclamation and closure regulations. EPA chose subjective, inflexible, and arbitrary criteria with no scientific or other rational explanation for why they were chosen. For example, some of the reduction criteria are cherry picked from state regulations (e.g., New Mexico) but completely fail to take into account local considerations, such as geology, climate, seismology and ecology. Moreover, even those that are cherry picked exclude important caveats and additional regulatory language that would make reduction criteria inappropriate in certain circumstances or directly contradict state regulations. In many cases the conditions for a reduction are so highly prescriptive that they can be met only through specific engineering controls which will lead to such controls being adopted even where they serve no real-world purpose in reducing risk. Further, EPA's chosen reduction criteria are likely to provide no practical regulatory relief because they are embedded in an "all or nothing" approach - a mine would get 100% credit or 0% credit. EPA's own RIA shows that only one of the 49 facilities modeled received 100% reductions. See U.S. EPA, RIA at B-21. c. EPA's extrapolation compounds its flawed analysis. While EPA essentially admits that the modeled universe is not representative of the full universe, it nevertheless extrapolated data from the modeled universe to assess the impact on the universe of facilities that will be regulated. The extrapolation process, through multiple actions, further compounded the indefensible cost estimates. These actions included: EPA selecting of, without justification, the median cost rather than the mean cost of the modeled universe as the appropriate measure to use for extrapolating the results. See 82 Fed. Reg. 3392. Using a mean would more than double the estimated cost to industry and result in an even larger cost/benefit disparity. See RIA, Appendix F. EPA assuming that those who can self-guarantee likely would do so and extrapolating the results to the full universe. This extrapolation seems unlikely because EPA only had financial information for publicly traded companies, and privately held and smaller companies will likely have a harder time satisfying the stringent financial test. RIA at 5-6. EPA applying locality adjustment factors to account for regional variation in labor and material costs but failing to address how these locality adjustment factors then impacted the extrapolation. 82 Fed. Reg. 3391. EPA applying risk reductions in a way that it acknowledged involved oversimplification and required that it recognize there are "uncertainties" with the approach. 82 Fed. Reg. 3391. EPA assuming the net cost to the owner or operator of acquiring funds is the weighted average cost of capital ("WACC"), which it collected from the 21 firms' websites. 82 Fed. Reg. 3392. EPA makes no meaningful effort to claim or defend that the average WACC used reflects reality for the remaining 83% of facilities. Sierra Club v. EPA 18cv3472 NDCA Tier 7 ED 002061 00086796-00016 Docket No. EPA-HQ-SFUND-2015-0781 July 11, 2017 Page 17 d. ERA'S assessment of impacts is inadequate. EPA then presents a "screening-level" assessment regarding potentially significant impacts. This is a flawed approach in several respects. 82 Fed. Reg. 3395. First, for a rule of this magnitude, a screening level assessment is insufficient to support the Proposed Rule's conclusions. Second, as noted above, the modeled universe does not supply representative data. In addition, EPA assumes there are no collateral or annual costs in obtaining a financial responsibility instrument. For the firms within the modeled universe with cash flow under $1 million, the annualized costs of financial responsibility relative to cash flow is over 160%, with 5% of companies having impacts of over 10%. 82 Fed. Reg. 3395. In the end, EPA concludes that, consequently, this analysis is completely useless: Due to limitation in the financial data, EPA did not expand the screening analysis to the full universe of regulated facilities. EPA acknowledges that the results generated based on the modeled universe may not be reflective of the impacts on the entire industry. 82 Fed. Reg. 3395. Thus, EPA's economic analysis of the Proposed Rule can only be used to demonstrate that the cost of compliance relative to cash flow will be devastating to many companies, and the cost to industry far outweighs the EPA assumed regulatory benefit. Such a rule is the height of arbitrary and capricious. EPA similarly fails to present any useful employment impact analysis. Despite recognizing the requirements of Executive Order 13563 to conduct such analysis, EPA "did not have sufficient data to model and quantify the potential changes in mines' employment levels as a result of the proposed regulation." 82 Fed. Reg. 3395; RIA at ES-13. Thus, while EPA showed itself more than willing to speculate about potential benefits and discuss such benefits qualitatively, EPA refuses to do so regarding employment impacts on the other side of the ledger. The reality is that the staggeringly high costs of the Proposed Rule will have real economic impact - substantially impacting existing domestic jobs and investment and, consequently, the communities where mines are located; contracting the domestic mining market; putting American companies at significant global market disadvantage; and diverting productive investment. In a huge understatement, EPA acknowledges only that "it is possible that increased compliance costs will cause individual mines to adjust their employment levels to maintain overall profitability." RIA at 6-6. Not surprisingly, EPA cannot certify that there is no Significant Impact on a Substantial Number of Small Entities. RIA at ES-13; see also Johnson Declaration at 1135. In fact, the United States Small Business Administration submitted comments on the Proposed Rule, concluding that the Proposed Rule would "impose costly requirements on hardrock mines owned by small firms, without evidence that a problem exists warranting intervention." Advocacy Comments at 1; see also RIA at 8-2. The Small Business Administration Office of Advocacy indicates that 36% of hardrock mining businesses are small businesses, Advocacy Comments at 2, and EPA's own estimates indicate that this single regulatory Sierra Club v. EPA 18cv3472 NDCA Tier 7 ED 002061 00086796-00017 11 Docket No. EPA-HQ-SFUND-2015-0781 July ,2017 Page 18 requirement would impose costs in excess of three percent of revenue for many of these small mines. Advocacy Comments at 3. Further, EPA's one-to-one transfer of the industry costs as a social benefit to the financial sector is irrational because the real social impact is different. The productivity of the funds in the hands of mining and processing companies, which create jobs and products, cannot be compared to the passive function of the financial industry that will serve as issuers or guarantors of financial instruments. Overall, the Proposed Rule runs directly contrary to the Administration's regulatory reform agenda as embodied in various recent Executive Orders,11as well as other long-standing Executive Orders that seek to foster sound cost-benefit analysis.12These orders generally insist on an effective and cost-beneficial regulatory approach that avoids unnecessarily draining resources from private enterprises, instead directing those resources towards beneficial economic use. The Proposed Rule is a text book example of what those orders seek to avoid. V. EPA failed to follow the statutory mandate to consult with the financial industry to the maximum extent possible. Section 108(b) requires, "[t]o the maximum extent practicable, the President shall cooperate with and seek the advice of the commercial insurance industry in developing financial responsibility requirements." 42 U.S.C. 9608(b). This mandate was emphasized and expanded by Congress in its 2014 EPA appropriations bill. H.R. 2279,113th Cong. 103-105 (engrossed in H.R. Jan. 9, 2014). There Congress directed EPA to "collect and analyze information from the commercial insurance and financial industries regarding the use and availability of necessary instruments . . . for meeting any new financial responsibility requirements." Johnson Declaration 1)31. EPA acknowledges numerous unusual aspects to any 108(b) program and the financial instruments it has proposed, aspects which could greatly impact market availability, to say nothing of cost. Among these unique features are the large number of creditor classes that would be given access to the instruments, the lack of a clear mechanism to release financial instruments (even upon transfer of the facility), and the right of direct action against an instrument. This last feature alone presents a potential host of legal and practical problems that will insure such instruments will be very costly to the regulated entities. See 82 Fed. Reg. 3412-14. Even in the limited consultation EPA undertook, this led to "[representatives of the financial community expressing] differing levels of comfort regarding these unique aspects of the potential instruments under the EPA's CERCLA 108(b) rulemaking." MCS at 5. 11See, e.g., Executive Order 13777 of February 24, 2017, Enforcing the Regulatory Reform Agenda (directing agencies to identify regulations that eliminate jobs, are outdated, unnecessary or ineffective or that impose costs exceeding benefits); Executive Order 13771 of January 30, 2017, Reducing Regulation and Controlling Regulatory Costs ("it is essential to manage the costs associated with the governmental imposition of private expenditures required to comply with Federal regulations"). 12See, e.g., Executive Order 12866 of October 4,1993, Regulatory Planning and Review ("each agency shall assess both the costs and benefits of the intended regulation and . . . adopt a regulation only upon a reasoned determination that the benefits of the intended regulation justify its costs"). Sierra Club v. EPA 18cv3472 NDCA Tier 7 ED_002061_00086796-00018 Docket No. EPA-HQ-SFUND-2015-0781 July 11, 2017 Page 19 The Market Capacity Study was conducted too early in the process - before the formula was even developed - to allow meaningful input. Such timing assured that EPA could not satisfy the statutory mandate to consult with the industry to the maximum degree practicable, thus preventing the Market Capacity Study from serving as a sound basis on which the Proposed Rule can be justified as one that will be practicable and effective. The Study contains only cursory analyses focusing primarily on the insurance market in general, rather than environmental risk insurance specifically. It also reviews market capacity for instruments for which the markets are familiar, rather than adequately considering the impacts of the unique instrument features on the available market. Even with review of the traditional framework, EPA recognizes that it is "exceedingly difficult to make inferences or predictions from the data as to future market trends." MCS at 8. That is why consultation with the industry is so important and why it is critical to move beyond published data in order for EPA to meet its statutory duty to consult. Yet, the Study was conducted without significant input from the financial industry. Ultimately, EPA has failed to demonstrate that appropriate capacity will be available. First, EPA limited itself to published industry data regarding the existing market for insurance and surety instruments. These data are not sufficiently instructive given the expansive scope and unique instrument features that EPA proposes to adopt under the Proposed Rule. Tellingly, EPA concedes that its conclusion is "highly dependent upon the overall amount of financial responsibility that the market will need to accommodate." 82 Fed. Reg. 3399. On the one hand, EPA "assumed that no market capacity constraints exist for the issuance of " third-party instruments." 82 Fed. Reg. 3392. Yet, at the same time, EPA recognizes that "tgjiven the number of unknown factors, the ultimate availability of CERCLA 108(b) financial responsibility instruments cannot be predicted with certainty until the final rule has been promulgated." 82 Fed. Reg. 3399. In the Regulatory Impact Analysis, EPA is even more direct: "EPA determined that the market for the types of FR instruments [to be issued by the insurance industry] does not vet exist to cover financial responsibility under CERCLA 108(b)." RIA at 4-10 (emphasis added). Given that the market does not currently exist, and the number of unknown factors EPA has acknowledged, it is clear that the Proposed Rule lacks a solid foundation on which to move forward. EPA's confidence that sufficient market capacity will be present is also directly at odds with its acknowledgement that the insurance and surety markets for environmental liabilities are contracting. MCS at 2. EPA ultimately appears to be pinning its hopes on the development of the Alternative Risk Transfer market, specifically Risk Retention Groups, MCS at 2-3. The Proposed Rule, however, does not specifically allow access to these markets and they only remain under consideration by the Agency. 82 Fed. Reg. 3399. A regulation that assumes a market solution that the Proposed Rule itself would not allow cannot be sound. EPA also recognizes that, given the contraction of these markets, "a range of diverse financial assurance instruments may be needed to help ensure sufficient coverage for entities potentially regulated under any CERCLA 108(b) rules." MCS at 3. The Proposed Rule, in contrast, provides a limited range of very constrained instruments.13 Equally unsound is banking a proposal on 13Among the options that have been overlooked or ignored that would affect financial assurance or its costs are (a) Sierra Club v. EPA 18cv3472 NDCA Tier 7 ED_002061_00086796-00019 Docket No. EPA-HQ-SFUND-2015-0781 July 11, 2017 Page 20 the hope that the market will respond over time, while conceding the large uncertainties embedded in that hope. See, e.g. MCS at 5 ("it is not possible to predict the exact market for these instruments."). Even while acknowledging its inability to predict the "exact market," EPA understates its limits. It is not simply the "exact market" but the market writ large that EPA has failed to study sufficiently to offer any meaningful prediction. The huge uncertainties in this market are undeniable. For example, AIG will no longer offer environmental impairment liability coverage. AIG is the largest underwriter of environmental insurance for large scale, long duration environmental risk. MCS at 15. Further, the Market Capacity Study concludes that the marketplace "sh[ies] away from volatile lines of coverage." MCS at 21. Moreover, it is evident that where it is impossible to meaningfully predict the likely available capacity, it is an even greater leap of faith to predict the premiums to be charged for that capacity. And that premium uncertainty must in turn feed directly into greater uncertainty of EPA's cost estimates. The Market Capacity Study did not discuss trust funds because they do not "directly [rely] upon the availability of third-party markets for financing" and thereby concluded there is unlimited market capacity. MCS at 1-2. This conclusion, however, is unsupportable given the unique requirements of the financial instruments derived from either the statute or the Proposed Rule, particularly the direct action provision. In short, while the ability to fund a trust fund may depend solely on one's credit-worthiness, the ability to identify a third-party trustee willing to take on the risks inherent with the proposed instrument's structure is a different story. This is not simply ungrounded speculation. EPA reports that, with the exception of insurance providers, the financial instrument providers expressed "some degree of aversion" to the direct action provision, expressing concerns over invalid claims and legal fees. 82 Fed. Reg. 3414. Similarly, letter of credit providers stated that the direct action provision would be "out of the realm" of typical bank allowing captive insurance where the financial stability of a captive entity is separate and distinct from its parent, can be assessed and rated on a separate basis and is subject to its home state's insurance regulatory department; (b) alternatives to standby trust funds; and (c) eliminating requiring standby trusts where an insurance policy is present. Captive insurance may offer an alternative for entities that produce commodities and products whose value fluctuates greatly in the market. Certain companies can face challenges in accurately predicting whether they will meet the financial ratio criteria for issuing a corporate guarantee at all times, in all market conditions. Thus, for these, captive insurance can be a more attractive financial responsibility mechanism, as compared to the financial test and corporate guarantee mechanisms. A captive holds a defined group of assets--the capital and surplus to cover the risks it insures--and under bankruptcy laws, the captive's assets must be used first to pay claims that are made on the policies it has written. If the captive has surplus assets left over after all such claims are paid, then, and only then, will those surplus assets become available to the captive's parent company, and used to pay other creditors in the event of a parent company bankruptcy proceeding. As to requiring standby trusts where insurance or another financial instrument is available, a series of standby trusts merely adds thousands of dollars of additional costs each year, and no good purpose or environmental benefit is served by requiring owners and operators to pay a trustee to establish and hold an empty trust account for decades. If an insurance policy, letter of credit or bond is triggered, and if a trust fund is needed to hold the proceeds, then a trust fund can be set up at that time. These recommendations are merely meant to maintain maximum flexibility of third-party instruments available if the agency were to finalize this rule or any other rule under CERCLA 108(b). It is not meant to suggest that it can solve the inherent market capacity issues {e.g., availability and affordability) presented in this rulemaking. Sierra Club v. EPA 18cv3472 NDCA Tier 7 ED 002061 00086796-00020 Docket No. EPA-HQ-SFUND-2015-0781 July 11, 2017 Page 21 responsibilities, and banking institutions that serve as trustees stated that trust institutions would not participate in a program where the institution can be subject to any liability. MCS at 7. Thus, EPA's own record undermines its companion claims that: (1) the establishment of trust funds and procurement of letters of credit depend primarily on creditworthiness; and (2) the market is "essentially unlimited." Ultimately, because EPA did not seek advice from the financial industry "to the maximum extent possible," or even to any reasonable amount, EPA has been left to guess at what the financial industry would find palatable. See 82 Fed. Reg. 3415. This has left the Agency in the position of seeking, via the Proposed Rule comment process, the feedback that should have preceded the Proposed Rule's announcement, as required by statute. 82 Fed. Reg. 3414. Finally, and amazingly given all the uncertainty regarding market capacity, EPA has proposed eliminating the Financial Test and Corporate Guarantee, even as it concedes that including a financial test will help alleviate pressure on third-party surety markets to ensure greater market capacity, 82. Fed. Reg. 3399, and would provide significant cost savings. 82 Fed. Reg. 3431. As an initial matter, this is contrary to the plain language of the statute, which states, "[fjinancial responsibility may be established by any one, or any combination, of the following: insurance, guarantee, surety bond, letter of credit, or qualification as a self-insurer" (emphasis added). 42 U.S.C. 9608(b)(2). Thus, EPA can promulgate only a Proposed Rule that affirmatively includes a self-insurance option. Second, even absent the statutory language, the financial test is a proven successful option for financial responsibility. Including a financial test and corporate guarantee will have the benefit of easing the anticipated market constraints. Moreover, while EPA suggests concerns about using the financial test in the HRM sector, EPA has set for itself a higher bar for determining whether a financial test should be available, given its claim that these regulations will provide the model and will become applicable to any later sector for which 108(b) financial responsibility is required. Among the issues that were not adequately considered are the test thresholds that EPA has proposed, which are unnecessarily strict, such as linking use of the financial test to a higher than investment grade rating. This is particularly a concern when considering other industries that may become subject to these regulations given the EPA's failure to calibrate these levels with any application other than the HRM industry in mind. Every aspect of EPA's failure to adequately consult with and understand the financial markets will be seriously exacerbated if EPA moves forward to regulate other industrial sectors under 108(b). Limited availability and high costs of such instruments will have a devastating impact on regulated industries and, correspondingly, the United States economy. VI. The rulemaking conducted by EPA was significantly flawed. a. EPA's rush to propose a rule resulted in an indefensible rule developed without adequate stakeholder input. The Proposed Rule's flaws flow, at least in part, from its rushed development. EPA did not allow appropriate time for study, deliberation, consultation with important stakeholders, such as industry and Sierra Club v. EPA 18cv3472 NDCA Tier 7 ED 002061 00086796-00021 Docket No. EPA-HQ-SFUND-2015-0781 July 11, 2017 Page 22 state and federal regulators, and reasoned consideration of the costs, benefits and consequences of the rules. While EPA maintained adoption of a final rule by August 2019 would be "aggressive," Johnson Declaration at H 73, the agency negotiated behind closed doors with the petitioners an even more expedited timeframe for a final rule by January 1, 2017. See In re Idaho Conservation League, No. 14-1149 (DC Cir. January 29, 2016). Given EPA's greatly accelerated and aggressive rulemaking schedule, EPA did not have an adequate opportunity to properly consider the need for the rule, let alone develop a rule that is based on all appropriate consideration and cost/benefit analysis. Indeed, as predicted by Barnes Johnson in 2014, the abbreviated timeframe has prevented EPA from developing a "carefully crafted proposal" due to "procedural [and] analytical shortcuts that [] jeopardize the soundness of the regulations, their legal defensibiiity [], and raise the potential for unintended consequences on state financial assurance programs and for the regulated community." Johnson Declaration at 1137. EPA now presents a Proposed Rule without appropriate study or deliberation, without necessary stakeholder participation, and without heeding the striking cost/benefit imbalance that its own truncated analysis demonstrates. The determination of the "degree and duration of risk" associated with any industrial sector is complex and requires significant data and analysis. See Johnson Declaration at HH6,12,16, 33. EPA rebuffed many offers of assistance by the HRM industry to supply operational data, ignored information provided by affected states regarding the comprehensive regulatory programs already in place, and barely engaged with the financial industry regarding market capacity and appropriate instruments. This approach suggests a pre-conceived bias and lack of fairness in the EPA assessment. Because there has been little stakeholder input, the Proposed Rule relies on limited information that is not complete or representative. The WGA Comments and Advocacy Comments describe the lack of substantive consultation, EPA's failure to provide the most basic of information, and EPA's complete disregard of concerns raised regardless of the requirements to address federalism and small business impacts. See August WGA Comments, March WGA Comments, Advocacy Comments; see also Executive Order 13132 (August 4,1999) and Small Business Jobs Act of 2010 (P.L. 111-240). Similarly, the rush to publish the Proposed Rule by the court-ordered deadline compromised EPA's peer review process, which fell far short of the requirements in OMB and EPA peer review guidelines. See U.S. Office of Management and Budget, Final Information Quality Bulletin for Peer Review, M05-3, December 16, 2004; U.S. EPA, Peer Review Handbook 4th Edition (EPA/100/B-15/001) (October 2015). The process was extremely limited, reviewing, for example, limited aspects of only the formula. The reduction criteria and reports on "continuing risk," among other things, were not peer reviewed. The process also was rushed and completed only a short time before the proposal was issued, raising serious doubts that the peer review comments were substantially considered before the Proposed Rule was signed. Notwithstanding this, even with the limited and abbreviated peer review and inadequate time for proper inclusion, the only conclusions that can be drawn from the results of the peer review process is that the Proposed Rule must be withdrawn. Sierra Club v. EPA 18cv3472 NDCA Tier 7 ED 002061 00086796-00022 Docket No. EPA-HQ-SFUND-2015-0781 July 11, 2017 Page 23 b. The rule violates the Information Quality Act. Congress enacted the Information Quality Act (IQA) in December 2000 as Section 515 of the Treasury and General Government Appropriations Act for Fiscal Year 2001. (Public Law 106-554). The Act required the Office of Management and Budget (OMB) to issue guidance to federal agencies designed to ensure the "quality, objectivity, utility, and integrity" of information disseminated to the public. It also required agencies to issue their own information quality guidelines. In short, the IQA, OMB guidelines and EPA guidelines require that information be accurate, reliable, and unbiased, and presented in a complete and unbiased manner. See e.g., Office of Management and Budget, Guidelines for Ensuring and Maximizing the Quality, Objectivity, Utility, and Integrity of Information Disseminated by Federal Agencies; Notice; Republication, 67 Fed. Reg. 8452 (February 22, 2002) ("OMB Guidelines"); United States Environmental Protection Agency, Guidelines for Ensuring and Maximizing the Quality, Objectivity, Utility, and Integrity of Information Disseminated by the Environmental Protection Agency, EPA/260R-02-008 (October 2002, as revised June 24, 2004 and May 13, 2005) ("EPA Guidelines"). There are additional requirements for scientific information that is deemed to be "influential"; i.e. information that has a clear and substantial impact on important public policy or private citizens. Such influential information must also be "reproducible" according to commonly accepted scientific, financial, or statistical standards so that a third party could conduct the same analysis and arrive at substantially the same result. EPA Guidelines at 21. These standards apply not only to the substance of the information but also to EPA's analysis of that information and its presentation. As demonstrated herein, the Proposed Rule clearly violates these requirements, and this violation makes finalizing the Proposed Rule unlawful. Pursuant to EPA's guidelines, information includes any communication or representation of knowledge such as facts or data in any medium or form that, when disseminated, must meet "a basic standard of quality," defined in terms of objectivity, utility and integrity. EPA Guidelines at 3,15. EPA disseminates information when it prepares and distributes information to, among other things, formulate or support a regulation. EPA Guidelines at 15. "'Objectivity' focuses on whether the disseminated information is being presented in an accurate, clear, complete, and unbiased manner, and as a matter of substance, is accurate, reliable and unbiased." EPA Guidelines at 15. Influential information must have a higher degree of transparency that will facilitate reproducibility according to commonly accepted scientific, financial or statistical standards. EPA Guidelines at 20-21. The presentation must be comprehensive, informative and understandable, and must specify: (1) each population addressed; (2) the expected risk or central estimate of risk for each population; (3) each appropriate upper- or lower-bound estimate of risk; (4) each significant uncertainty and studies that would assist in resolving the same; and (5) peer-reviewed studies that support, are directly relevant to, or fail to support estimates and methodologies used to reconcile inconsistencies in data. EPA Guidelines at 22-23. Sierra Club v. EPA 18cv3472 NDCA Tier 7 ED_002061_00086796-00023 Docket No. EPA-HQ-SFUND-2015-0781 July 11,2017 Page 24 The EPA Guidelines further incorporate by reference the EPA Peer Review Policy. Guidelines at 11; see also United States Environmental Protection Agency, Peer Review and Peer Involvement at the USEPA (June 7,1994); United States Environmental Protection Agency, Science and Technology Policy Council, Peer Review Handbook, 4th Edition, EPA/100/B-15/001 (October 2015). Among other things, the Guidelines indicate every effort should be made to complete the peer review prior to the proposal stage. Peer Review Handbook at 28. Furthermore, peer review should be transparent and be conducted in an open and rigorous manner. 67 Fed. Reg. 8454, 8459-60. The studies and analyses purported to support the Proposed Rule do not meet these standards. The private, incomplete peer review that is discussed above provides more than adequate evidence of this failure. Not only does the peer review not satisfy the requirements because neither a compliant review was completed in advance of the proposed rule nor conducted in a transparent manner but, based on the comments from the reviewers, the data are not accurate, reliable, unbiased or reproducible. As discussed above, Peer Review Commenter 4 could not replicate a single cost number when conducting spot checks, and "expressed concern about the procedure used to collect cost and acreage data from the reclamation and closure plans." Commenter 4 suspects there are errors that result from a misunderstanding of the information in the original source documents and argued that EPA should redo the data collection using mining industry specialists. Response to Peer Review Comments at 2-8 - 2-9. Reviewers had concerns about sample and data representativeness, data integrity, sources of data, statistical methods, regression error, and model accuracy, among other things. Response to Peer Review Comments at 2-1, 2-8 to 2-10, 2-12, 3-1 to 3-4, 3-10 to 3-11, 3-15 to 3-19. In short, three of the four Peer Reviewers "suggested EPA provide a more detailed description of the data and how it was used to construct the formula in order to improve transparency." Response to Peer Review Comments at 6-1. The OMB Guidelines mandate that agencies establish, "[a]s a matter of good and effective agency information resources management," a "pre-dissemination review" process. Under this process, agencies are to review (and substantiate) the quality of information at every step in its development, from creation, collection and maintenance to dissemination. The EPA Guidelines incorporate this pre-dissemination review process, but there is no evidence that EPA complied with that process in this case. To the contrary, as discussed above, there is significant evidence that EPA developed the Proposed Rule behind closed doors and refused to disseminate information to affected parties including affected states, the Small Business Administration and potentially regulated entities. Based on the foregoing, the following conclusions can be made: First, the quality requirements established by the IQA, OMB Guidelines, and EPA Guidelines apply to information disseminated by EPA in all cases, regardless of whether a correction request is ultimately filed. That is the point of the pre-dissemination process. Sierra Club v. EPA 18cv3472 NDCA Tier 7 ED_002061 00086796-00024 Docket No. EPA-HQ.-SFUND-2015-0781 July 11, 2017 Page 25 Second, failure to meet these requirements in all cases will make agency action "unlawful1' as "not in accordance with law" or "without observance of procedure required by law." Third, failure to use objective information in a rulemaking is a failure to exercise reasoned decision-making, and hence is arbitrary and capricious. Finally, failure to present information in a rulemaking in an objective way is arbitrary and capricious. This result is consistent with courts' construction of the APA as requiring that an agency "make its views known to the public in a concrete and focused form so as to make criticism or formulation of alternatives possible." c. The Proposed Rule is so riddled with errors, it is neither factually nor legally defensible. The flawed analyses and conclusions in the Proposed Rule are not limited to EPA's failure to properly assess risk, conduct defensible cost estimates, develop a formula based in reality or identify financial instruments that would be palatable and available to financial markets. Rather, the Proposed Rule has many provisions that are arbitrary and capricious or so vague in their terms and meaning that the Proposed Rule is rendered invalid, even without EPA's failure to comply with the statute's direction to assess the "degree and duration of risk" and consult with the financial industry. For example, as noted above, the proposed reduction methodology to reduce the required financial responsibility amount is based on a review of a limited number of programs and a weak understanding and misapplication of existing financial responsibility programs. As another example, for a reduction in financial responsibility amounts, an owner or operator would have to demonstrate that it is subject to, and in compliance with, requirements that will result in a "minimum" degree and duration of risk. 82 Fed. Reg. at 3505; Proposed Section 320.63(c). EPA fails, however, to provide any information on what would constitute a "minimum" degree and duration of risk and how that would be demonstrated. Similarly, to be released from the financial responsibility obligations, the owner or operator would have to submit "evidence" demonstrating the degree and duration of risk to be "minimal." 82 Fed. Reg. at 3489; Proposed Section 320.27. Upon receipt, EPA will evaluate it and make a determination. However, EPA offers no explanation of what "evidence" is required, what constitutes "minimal" risk, or in what time period a decision will be made, leaving the regulation excessively vague. Yet another example of a fatal defect in the Proposed Rule is EPA's approach to the natural resource damages ("NRDs") component of the formula. First, inclusion of NRDs at all is in excess of EPA's statutory authority. EPA's 108(b) authority is directed at insuring that clean-up costs are met NRDs are not such costs. Under CERCLA the President has delegated authority to EPA to ensure that hazardous substance response costs are paid by the parties responsible for releasing them into the environment. Exec. Order No. 12580, 52 Fed. Reg. 2923 (Jan. 23,1987); 42 U.S.C. 9601 etseq. NRDs are not response costs under CERCLA Sections 101(6) and (25). 42 U.S.C. 9607(6), (25). Including them within the formula for required financial responsibility thus unduly expands EPA's authority Sierra Club v. EPA 18cv3472 NDCA Tier 7 ED_002061 00086796-00025 Docket No. EPA-HQ-SFUND-2015-0781 July 11, 2017 Page 26 beyond its current limitations. Moreover, EPA is not an authorized claimant for NRDs arising from the HRM industry, and EPA cannot demand or collect a bond for a damage claim that it is not legally authorized to bring. Including NRDs within the financial responsibility amount that must be provided unlawfully expands the scope of 108(b). Additionally, the Proposed Rule attempts to enable a variety of other persons who are not authorized to bring NRD claims to demand and collect payment from CERCLA 108(b) financial responsibility for those damages. EPA's proposed rule improperly seeks to allow other government agencies and third-party CERCLA claimants to recover NRDs in situations where those entities are not authorized to bring such claims under CERCLA 107(f)(2) and have no legal obligation to use the recovered funds to restore natural resources for the public's benefit. Under CERCLA 107(f), appointed natural resource trustees are the only persons with the authority to bring NRD claims, recover such damages, and use the funds to restore, replace, or acquire equivalent natural resources. 42 U.S.C. 9607(f). Congress clearly limited the ability to pursue NRD claims to those natural resource trustees that are expressly identified as such by the President, the Governor of each State, or recognized Indian Tribes, under CERCLA Section 107(f)(l)-(2). 42 U.S.C. 9607(f)(l)-(2). Congress further limited the use of all sums recovered by an authorized natural resource trustee "for use only to restore, replace, or acquire the equivalent of' the injured resources under CERCLA Section 107(f)(1). 42 U.S.C. 9607(f)(1). CERCLA does not require, and in fact prohibits, the payment of NRDs to persons who are not designated natural resource trustees and who have no legal obligation to use the recovered funds to restore resources for the public's benefit. Even if including NRDs were legally defensible, the generalized approach to estimating NRDs is indefensible, given the site-specific nature of NRDs. First, because EPA modeled the NRD costs as a percentage of response costs, all the concerns and issues associated with the response cost formulas are transferred and present in EPA's NRD cost estimates. In addition, NRDs are driven by the resources proximate to a site and the remedy implemented. Any justification of the NRD approach would have to be based on a more accurate and robust data set than what the Agency has relied upon; e.g., the data would include sites for which there have been zero NRD costs and consider only HRM facilities. Similarly, in developing the NRD multiplier, EPA relied on data from facilities whose NRD estimates with vastly different operating profiles, including steel facilities and chemical processing plants. EPA's use of data across unrelated and disparate industries contradicts the statutory purpose of the formula imbedded in the Proposed Rule: to create a sector-specific cost estimate that fairly reflects the sector. Finally, the use of the mean versus the median data point cannot be justified. See, e.g., Response to Peer Review Comments at 4-10. EPA also fails to recognize the interrelation between achieving an effective clean-up and reducing NRDs. A more extensive remedy will greatly reduce or potentially eliminate NRDs. EPA itself acknowledges that NRDs and response costs are dependent on each other, but its formula does not reflect this fact. During the remedy selection phase of CERCLA, responsible parties will, among other things, look at the amount of natural resource injuries - if any - that will remain after a particular remedy Sierra Club v. EPA 18cv3472 NDCA Tier 7 ED 002061 00086796-00026 Docket No. EPA-HQ-SFUND-2015-0781 July 11, 2017 Page 27 is constructed or completed. If a less extensive remedy that is protective of human health and the environment is likely to leave natural resource injuries unaddressed, responsible parties can estimate the value of the likely NRD claim for those injuries, and consider whether a different remedy would reduce or avoid natural resource injuries and whether that remedy would provide a more cost effective solution for the site. EPA's simplistic approach to address NRDs cannot account for this inter-dependency, nor has EPA even attempted to do so. Instead, under EPA's multiplier approach, companies with the most extensive work plans are asked to post the greatest amount of financial responsibility for NRDs. The formula does not consider the possibility that a more extensive work plan will greatly reduce or eliminate potential natural resource injuries and claims, or vice versa. As a result, the amount of financial responsibility that EPA proposes to require for natural resource injuries may have little or no relationship to actual injuries, if there are any. Put another way, there is no cause to demand financial responsibility to cover hypothetical NRDs that may never arise because extensive remedy work is undertaken, yet EPA's working assumption in the Proposed Rule is that the Proposed Rule will insure that extensive remedies will occur and high NRDs with them. There is thus an inherent tension in including NRDs within the financial responsibility amount required. In sum, the NRD component has irredeemable flaws. VII. Conclusion As set forth above, the Proposed Rule far exceeds CERCLA's statutory directive for EPA to require financial responsibility that is limited to that which is consistent with the "degree and duration of risk" associated with the production, transportation, treatment, storage or disposal of hazardous substances. While there is no question that historical industrial operations and the lack of regulatory control in the early to mid-20thcentury resulted in an unfortunate legacy of contaminated sites, HRM and industrial manufacturing operations since that time have changed drastically, as has their surrounding regulatory regime. Corporate America's modern business practices have evolved to embrace strong internal environmental compliance programs and corporate responsibility metrics. Yet, EPA has proposed a regulatory scheme that rests almost exclusively on past industrial practices and assumes all aspects of every operation will fail and that no owner or operator will have any funds to address those releases. In doing so, EPA exaggerates the risk profile of the HRM industry and risks cementing an approach to assessing risk that will similarly exaggerate the risk profiles of other industry sectors. Compounding this exaggeration further, EPA greatly overstates the benefits associated with the Proposed Rule and greatly underestimates the costs. The Proposed Rule basis is further compromised by EPA's flawed analysis of financial markets and unsubstantiated hope that capacity will exist to assure this untested market. But, even taking EPA's clearly incorrect analysis at face value, the cost-benefit analysis cannot support the Proposed Rule. Facilities are better built, better run, and more tightly controlled and supervised as to all pollutants than in prior decades. Existing regulatory and oversight programs promptly address releases Sierra Club v. EPA 18cv3472 NDCA Tier 7 ED 002061 00086796-00027 Docket No. EPA-HQ-SFUND-2015-0781 July 11, 2017 Page 28 of hazardous substances and require proper closure and reclamation as assured through existing federal and state financial responsibility requirements. In turn, the Proposed Rule, or anything like it, will impose extraordinary costs on industry without any corresponding environmental benefit or assurance of avoided societal cost. It must be withdrawn. Respectfully submitted, Ronald J. Tenpas Sierra Club v. EPA 18cv3472 NDCA Tier 7 ED 002061 00086796-00028