December 1, 2014 Attn: Docket ID No. EPA-HQ-OAR-2013-0602 EPA Docket Center U.S. Environmental Protection Agency Mailcode: 28221 T 1200 Pennsylvania Ave., NW Washington, DC 20460 American Coalition for Clean Coal Electricity Comments on EPA’s Proposed “Carbon Pollution Emission Guidelines for Existing Stationary Sources: Electricity Utility Generating Units” The American Coalition for Clean Coal Electricity (ACCCE) submits the following comments on the Environmental Protection Agency’s proposed Carbon Pollution Emission Guidelines for Existing Stationary Sources: Electric Utility Generating Units (“proposed rule”). 1 ACCCE is a national trade organization comprised of industries involved in producing electricity from coal. 2 Coal-fueled power plants produce carbon dioxide (CO 2 ) emissions, which are the focus of this proposal. Therefore, ACCCE and its members have a direct and substantial interest in the proposed rule. In addition to our comments, ACCCE is a member of the Utility Air Regulatory Group (UARG) and incorporates the UARG comments on the proposed guidelines by reference herein. 1 For a host of reasons, ACCCE is opposed to regulating CO 2 emissions under the Clean Air Act (CAA or Act). In fact, this proposal demonstrates by its very nature why such regulation is a very bad idea. However, because EPA has proposed to regulate CO 2 emissions from existing power plants under section 111(d) of the Act, we are providing comments that demonstrate why EPA should abandon its proposal due to profound legal flaws, as well as substantial policy and technical shortcomings. Executive Summary On June 2, EPA released the Clean Power Plan (CPP), a propos al to regulate CO 2 emissions from existing fossil fuel-fired power plants. 3 In the proposal, EPA sets CO 2 emissions rate targets for each of 49 states. EPA’s state targets are based on EPA’s assumptions regarding emissions reductions that it estimates can be achieved through some combination of four “Building Blocks.” EPA projects that its proposal will cost up to $8.8 billion in 2030 and reduce power sector CO 2 emissions by 30 percent below 2005 levels by 2030. However, EPA’s proposal is profoundly flawed from a legal standpoint and should be withdrawn:  First and foremost, EPA has no authority to regulate CO 2 emissions from coal-fired power plants because the CAA prohibits source categories regulated under section 112 -- in particular, coal-fired power plants -from being regulated under section 111(d). However, EPA has proposed to regulate coal-fired power plants under section 111(d).  Putting aside EPA’s lack of authority to regulate at all (immediately above), in developing guidelines for states to use to set 111(d) 2 performance standards, EPA is allowed by the CAA to consider only emissions reductions that can be achieved at a regulated facility , i.e., “inside the fence.” However, EPA has proposed standards based on measures clearly “beyond the fence.”  EPA has not met the legal prerequisites for regulating existing plants under section 111(d) of the CAA.  Finally, EPA’s beyond-the-fence approach to setting standards impermissibly usurps states’ traditional sovereign power to regulate electric generation and use within their borders. Moreover, EPA has made incorrect and arbitrary assumptions concerning all four Building Blocks. These errors provide further compelling grounds for EPA’s withdrawal of the proposed rule:  For Building Block 1, EPA has significantly overestimated the potential for heat rate improvements at existing power plants by misinterpreting historical power plant operating data and an engineering analysis of power plant heat rate improvement. 4  For Building Block 2, EPA overestimated the potential increase in natural gas combined cycle (NGCC) capacity factors by ignoring natural gas infrastructure constraints and the historical role of NGCC power plants in serving intermediate load, not base load, electricity generation. 5  For Building Block 3, EPA overestimated the potential for renewable energy (RE) due to an arbitrary and incorrect application of state Renewable Portfolio Standards (RPS) to states without such standards. 6  For Building Block 4, EPA inappropriately derived a simplistic year-onyear growth rate for end-use energy efficiency (EE), ignoring the EE that 3 has already been deployed, and the different capabilit ies of the industrial, commercial and residential sectors to increase efficiency. 7 These and other errors clearly demonstrate the arbitrary and capricious nature of the proposed EPA guidelines and, as result, EPA should withdraw its proposal. In addition to being illegal and technically flawed, the CPP is the most expensive environmental regulation ever proposed for the power sector. 8 NERA Economic Consulting conservatively estimates that the CPP will cost at least $366 billion between 2017 and 2031, with an annual cost of at least $41 billion. 9 NERA further estimates that electricity prices will increase by an average of 12 percent nationwide, with 43 states experiencing double digit price increases. Furthermore, consumers are projected to pay -- directly out of their pockets and indirectly through higher electricity prices -- $560 billion to reduce their use of electricity. In a second scenario, under which states or EPA are unable for legal and practical reasons to rely on Building Blocks 3 and 4 (increased EE and RE), the cost of the CPP could reach $479 billion between 2017 and 2031, or $73 billion per year. Under this scenario, electricity price increases are projected to be even higher. The imposition of these exorbitant costs is not only harmful to our economy, but also underscores the illegality of any rule that imposes exorbitant compliance costs but, as discussed below, achieves no meaningful climate change benefits. 4 In addition, NERA projects coal retirements between 45,000 megawatts (MW) and 169,000 MW due to the CPP. Coal retirements, among other issues, have led organizations like the North American Electric Reliability Corpor ation, the Southwest Power Pool, the Midcontinent Independent System Operator, the Electric Reliability Council of Texas, and American Electric Power to express serious concerns about the CPP’s threat to electric reliability. 10 Despite the unprecedented costs and substantial risks to electric reliability, the CPP will have no meaningful effect on global climate change. For example, using EPA’s own methodology to estimate the climate effects of reducing CO 2 emissions shows that the CPP will reduce sea level rise by 0.03 millimeter and will reduce global temperature rise by 0.016 degree Fahrenheit in 2050. 11 In summary, EPA has no legal authority to take this action and, moreover, has based the state emissions targets on a series of incorrect and arbitrary assumptions. The proposal is exorbitantly costly, poses a serious threat to electric reliability, and will have no meaningful effect on global climate change. For these reasons, EPA must withdraw the proposed Clean Power Plan in its entirety. Sincerely, /s/ Paul Bailey Senior Vice President, Federal Affairs and Policy 5 Comments of the American Coalition for Clean Coal Electricity I. EPA Lacks Authority to Regulate CO 2 from Existing Coal-Fired Power Plants Under Section 111(d) of the Clean Air Act. A. The Statute Prohibits Section 111(d) Regulation Because EPA Has Regulated Coal- and Oil-Fired Power Plants Under Section 112 of the Clean Air Act. The Environmental Protection Agency (EPA) lacks authority to regulate CO 2 from fossil-fueled power plants under section 111(d) of the Clean Air Act (CAA or Act) because it has already issued final rules to regulate coal - and oil-fired power plants under section 112 of the Act. Under the plain terms of section 111(d), EPA may not regulate “any air pollutant … emitted from a source category which is regulated under section [112].” 12 EPA’s attempts to escape the plain language of the statute are not supported by the text and h istory of the CAA. The statute is not ambiguous on the question of whether EPA may regulate power plants under section 111(d), so EPA is not entitled to Chevron deference on this issue. 13 Even if, as EPA asserts, the existence of a clerical error indicates some ambiguity in the statute, and if EPA is entitled to resolve that ambiguity under the deferential Chevron principles of statutory construction, EPA’s proposed interpretation is not reasonable in light of the structure of sections 111 and 112 of the CAA and the clear purpose behind the 1990 CAA Amendments. 14 As stated above, the plain text of section 111(d) states that EPA may only require states to issue performance standards for pollutants that are not 6 “emitted from a source category which is regulated under section [112].” 15 EPA attempts to sidestep this clear prohibition by relying on a non-substantive “conforming amendment” found in section 302(a) of the 1990 CAA Amendments 16 – an amendment that the agency itself conceded wa s “a drafting error and therefore should not be considered.” 17 EPA’s reliance on this drafting error is misplaced for at least two reasons. First, EPA’s reliance on a non-substantive clerical error to reduce the effect of an amendment that is clearly substantive in nature contradicts bedrock principles of statutory construction. Second, even assuming that EPA may give effect to the non-substantive clerical error, EPA’s approach fails to give the maximum possible effect to the plain language of the supposedly conflicting provisions. Both courts and the compilers of the U.S. Code have long recognized that legislatures occasionally make clerical errors when rewriting complex sections of the legislative code. The U.S. Code contains dozens of examples of clerical errors that have been excluded from the U.S. Code in order to give effect to Congress’ true intent. 18 Courts have also long recognized that stray clerical errors should not be relied upon to defeat congressional intent. 19 Under the pre-1990 text of section 111(d), EPA was prohibited from regulating pollutants listed under section 112(b)(1)(A) of the CAA. The 1990 CAA Amendments changed the numbering of section 112 to eliminate section 112(b)(1)(A) entirely. As a result of the 1990 CAA Amendments, pollutants were to be listed under section 112(b). Section 302(a) of the 1990 CAA Amendments was a conforming amendment thus merely intended to update the statutory cross-reference in section 111(d) to reflect the revisions to section 112. It state d 7 “Section 111(d)(1) of the Clean Air Act is amended by striking ‘112(b)(1)(A)’ and inserting in lieu thereof ‘112(b)’.” In the case of section 111(d), it is clear that the conforming amendment contained in section 302(a) of the 1990 CAA Amendments was a clerical error and should be disregarded. First, this provision – which would have simply updated a stray reference in section 111(d) to reflect the renumbering of section 112 20 – is clearly labeled a “conforming amendment,” as it was in the Senate version of the 1990 CAA Amendments. The Senate Legislative Counsel Office’s Legislative Drafting Manual defines a conforming amendment as “a n amendment of a provision of law that is necessitated by the substantive amendments or provisions of the bill. The designation includes amendments, such as amendments to the table of contents, which formerly may have been designated as clerical amendments.” 21 The fact that both the Senate version of the 1990 CAA Amendments and the final text of the law label section 302(a) a “conforming amendment” indicates that this amendment was not intended to have any substantive effect on the law, but was merely included in earlier versions of the Senate bill to ensure that cross-references in section 111(d) would be updated to the revised structure of section 112. The fact that the inclusion of section 302(a) in the CAA Amendments of 1990 is a clerical error is reinforced by the observation that the amendment is rendered unnecessary by the substantive change to section 111(d) that Congress directed through section 108(g) of the 1990 CAA Amendments. That change, which deleted the reference in section 111(d) to the pollutant listing restriction of section 112 and replaced it with a restriction on regulating pollutants under Section 111(d) emitted from sources that EPA regulates under section 112, was 8 intended to reflect two substantive revisions that Congress made to EPA’s treatment of hazardous air pollutant (HAP) emissions. The first revision was to change the focus of section 112 from regulation of specific pollutants to regulation of source categories and to therefore eliminate EPA’s discretion over regulating HAPs from major sources. 22 The effect of this change was to dramatically increase the number of major sources and a rea sources that would be subject to emission limits for the 189 HAP compounds Congress listed in the statute, as well as any additional compounds EPA added to the list. 23 In effect, the expansion of 112 to ensure that all existing source categories that emit HAPs are regulated reduced the need to regulate existing sources of HAPS under section 111(d); hence, Congress logically added the exclusion from section 111(d) regulation for any source category regulated under section 112. The second revision had the effect of authorizing EPA to regulate HAPs from power plants under section 111(d) if – and only if – the agency concluded, after conducting a study of the effect of those emissions on human health – that regulating HAPs from power plants under section 11 2 would not be “appropriate and necessary.” 24 Because the pre-1990 CAA Amendments’ text of section 111(d) would have prevented EPA from regulating HAPs from power plants under section 111(d) if those HAPs were simply listed as pollutants under section 112, the 1990 CAA Amendments revised section 111(d) to clarify that EPA could regulate any existing source categories – including power plants – that the agency had decided not to regulate under section 112. 9 EPA itself has conceded that section 302(a) is likely a clerical error. After reviewing the legislative history of the 1990 CAA Amendments, EPA concluded in 2005 that “it appears that [section 302(a)] is a drafting error” and was not intended to be adopted by Congress. 25 In the CAMR rulemaking process, the agency erroneously concluded that it must nevertheless attempt to give effect to this non-substantive clerical error. 26 However, EPA’s decision to give effect to the clerical error is in error. Furthermore, as we explain below, this decision is especially egregious because EPA has relied on this clerical error to claim a broad new mandate for itself that was neither contemplated nor intended by Congress. B. Even Assuming that EPA Must Give Some Effect to the Clerical Error, EPA Lacks Authority to Regulate Power Plants Under Section 111(d). Even assuming that EPA is correct that it should consider the effect of the clerical error in determining its authority to regulate CO 2 , EPA is not entitled to deference unless it can demonstrate that the statute is ambiguous as to whether EPA may or may not regulate CO 2 from power plants under section 111(d), and that Congress “either explicitly or implicitly delegated authority to cure that ambiguity.” 27 EPA argues that the statute is ambiguous because the two provisions of the 1990 CAA Amendments conflict with one another “and thus render the Section 112 Exclusion ambiguous.” 28 EPA’s approach is wrong for at least three reasons. First, in jumping to the conclusion that the provisions are in conflict, EPA has ignored the canon of statutory construction that requires both courts and implementing agencies to 10 construe alleged conflicts in such a way as to give the maximum possible effect to every word in the statute. 29 Second, even if there were a conflict, EPA is not entitled to Chevron deference; rather, both EPA and the courts must attempt to discern the intent of Congress by looking to the usual tools of statutory construction. Third, even assuming that the conflict created an ambiguity, there is no indication that Congress intended to delegate to EPA the authority to cure the ambiguity. 30 EPA’s conclusion that the two provisions are in conflict, and thus, ambiguous, is based on a flawed premise. As the Supreme Court has often held, EPA and courts should not “lightly presume that Congress has legislated in self contradicting terms.” 31 When interpreting statutory language, EPA and the courts are required to “fit, if possible, all parts [of a statute] into [a] harmonious whole.” 32 In fact, sections 108(g) and 302(a) of the 1990 CAA Amendments can be read together as a “harmonious whole” that imposes complementary restrictions on EPA’s authority to regulate pollutants under section 111(d). Section 108(g) prohibits EPA from regulating pollutants (such as CO 2 ) that are emitted from source categories already regulated under section 112. Meanwhile, section 302(a) prohibits EPA from regulating pollutants that EPA lists under section 112. Although there may be significant overlap in terms of both sources and pollutants excluded from section 111(d), this overlap does not render this harmonious reading erroneous or ambiguous. Rather, these two provisions, when read together, give effect to both prohibitions by barring EPA from regulating both categories of pollutants (those emitted from source categories regulated under section 112 and those listed under section 112) . In addition, this reading would importantly not render section 111(d) moot. EPA would still be authorized to regulate non-HAPs under section 111(d) as long as those 11 non-HAPs were not emitted from a source category regulated under section 112 and were not being regulated as criteria pollutants under sections 108 and 109. EPA’s proposed interpretation, on the other hand, would give no effect to the section 108(g) prohibition on regulating pollutants emitted from sources regulated under section 112. Rather, EPA’s interpretation would allow it to regulate almost any pollutant emitted from a source category regulated under section 112 as long as the agency is not already regulating the specific pollutant under section 112. Because this interpretation fails to read allegedly “conflicting” provisions harmoniously by giving equal weight to both provisions of the 1990 CAA Amendments, EPA’s interpretation is invalid. EPA should interpret the arguably inconsistent amendments to section 111 as harmoniously as possible by giving full effect to both of the prohibitions that Congress included in the 1990 CAA Amendments. Moreover, even if there were an irreconcilable conflict between section 108(g) and section 302(a) of the 1990 CAA Amendments, EPA is not entitled to Chevron deference in resolving this conflict. The Supreme Court has held that agencies are not entitled to deference in resolving conflicts between otherwise clear statutory provisions unless the agency cannot “simultaneously obey” both provisions. 33 Here, EPA can “simultaneously obey” the prohibitions contained in sections 108(g) and 302(a) of the 1990 CAA Amendments. Finally, even assuming that the existence of two amendments to section 111(d) creates ambiguity, “the existence of ambiguity is not enough per se to warra nt deference to [an] agency’s interpretation. The ambiguity must be such as to make it appear that Congress either explicitly or implicitly delegated authority 12 to cure that ambiguity.” 34 Each of the amendments, standing alone, provides a clear, unambiguous statement about which pollutants EPA may regulate under section 111(d). In other words, neither amendment, standing alone, provides any indication that Congress meant to authorize EPA to cure an ambiguity over which pollutants the agency could regulate. It would be unreasonable for EPA to rely on the existence of two separate amendments—neither of which evinces any congressional intent to delegate authority to EPA to cure an ambiguity —as evidence that Congress intended to delegate such authority. There fore, EPA’s interpretation is not entitled to deference. C. Even Assuming that EPA Is Entitled to Chevron Deference, EPA’s Interpretation Is Unreasonable. Even if one concludes that EPA is entitled to Chevron deference, however, EPA’s interpretation is unreasonable. As the previous discussion demonstrates, EPA’s interpretation is unreasonable under Step 2 of Chevron because it begins with the assumption that Congress legislated in self-contradictory terms and fails to treat the 1990 CAA Amendments as a harmonious whole. Moreover, as the Supreme Court recently held in striking down another component of EPA’s greenhouse gas regulations under the CAA, “EPA’s interpretation is also unreasonable because it would bring about an enormous and transformative expan sion in EPA’s regulatory authority without clear congressional authorization. When an agency claims to discover in a long-extant statute an unheralded power to regulate ‘a significant portion of the American economy,’” the agency must provide an especially convincing explanation of its legal authority. proposed regulation under section 111(d) is far more 35 expansive EPA’s and transformative than the regulation at issue in Utility Air Regulatory Group v. 13 EPA. Here, EPA’s proposed interpretation would open the door to the regulation of CO 2 and other greenhouse gases from every category of existing stationary source under section 111(d). Such an interpretation would thus have “vast economic and political significance” 36 – a situation in which Congress would be expected “to speak clearly if it wishes to assign to an agency” such weighty decisions. 37 Yet, under EPA’s own theory of its authority under section 111(d), Congress has not spoken clearly on the question. 38 Rather, according to EPA’s argument, Congress legislated in such an unclear way that EPA has been forced to interpret congressional intent under Step Two of Chevron. Assuming that EPA is correct, and Congress did legislate in a conflicting or ambiguous manner, it is unreasonable for EPA to interpret a spare, little-used provision of section 111 – a provision that has been in effect for over forty years and which provides no hint that EPA may regulate either greenhouse gases or the entire electric power system – to somehow bestow on EPA the authority to impose costly, transformative regulations on a vast portion of the economy. Rather, assuming that the statute is ambiguous with respect to EPA’s authority to regulate CO 2 from existing power plants, the agency should interpret the statute in a way tha t does not require it to fashion an unprecedented, comprehensive regulatory scheme with the potential to significantly affect the entire U.S. economy. II. EPA Lacks Authority to Set State Emissions Goals Based on Outside-the-Fence Measures. EPA proposes to set CO 2 emissions rate goals for each state based on its interpretation of “best system of emission reduction” (BSER) for existing power 14 plants within the state. EPA’s interpretation of BSER includes “Building Block” actions that take place outside the fence line of the affected power plant and are outside of the control of the plant operator. This regulatory approach far exceeds EPA’s authority under the CAA for several reasons. First, the statute and previous EPA rules make it clear that the emissio n control systems that EPA may consider to be BSER include only those that can be implemented at and by the source. Consequently, EPA may not consider redispatch, RE, nuclear energy, or EE to be BSER because these measures are not part of an “inside-thefence” system of emission reduction that is within the source’s control. Second, EPA may not set the BSER based on redispatch from coal units to natural gas combined cycle (NGCC) because the CAA clearly requires continuous emission reductions – which would not necessarily occur under a redispatch approach to reducing emissions from coal-fired EGUs. Third and finally, EPA’s “alternative interpretation” of BSER—that reduced utilization constitutes BSER—is manifestly unreasonable and would represent an unprecedented departure from the intent and spirit of the Act. B. The CAA and Previous EPA Rules Are Clear: EPA May Only Consider Systems of Emission Control that Can Be Implemented at the Source. 1. Statutory Text The language of the CAA clearly indicates that EPA must determine BSER, and states must set performance standards on a source -specific basis, based on a system that is within the source’s control. Under section 111(d), EPA is authorized to direct states to establish “standards of performance for any existing 15 source . . . .” 39 As defined in section 111(a), a “standard of performance” means “a standard for emissions of air pollutants which reflects the degree of emission limitation achievable through the application of [BSER] which . . . the Administrator determines has been adequately demonstrated.” 40 Thus, the terms of section 111 – when read together – require that any eligible “system of emission reduction” be capable of “application” to the “existing sou rce.” This statutory linkage between emissions reduction systems to the existing source is a clear indication that EPA may only consider inside-the-fence systems of emission reduction in setting emissions guidelines under section 111(d). Moreover, because a standard of performance must be established “for any existing source,” each component included in the definition of “standard of performance” must also, logically, be “for any existing source.” Therefore, the “degree of emission limitation” on which the standard is to be based must be “achievable” “for any existing source.” Similarly, the “best system of emission reduction which … has been adequately demonstrated” must be a system that has been demonstrated “for any existing source.” A system canno t be considered “achievable” or “adequately demonstrated” for an existing source if the implementation of that system depends on the actions of third parties that are not under the source’s control. Two other provisions in section 111 reinforce the interpretation that the scope of BSER must be limited to “systems” of emission reduction that can be applied at the power plant unit itself. First, section 111(b)(5) of the CAA provides that, “except as otherwise authorized under subsection (h) of this section, nothing in this section shall be 16 construed to require, or to authorize the Administrator to require, any new or modified source to install and operate any particular technological system of continuous emission reduction to comply with any new source st andard of performance.” 41 This provision clearly speaks in terms of the performance standards that require sources to install and operate systems of emissions reductions within the fence line of the facility. While these reduction systems may generally be “technological” systems, section 111(b)(5) precludes EPA from requiring the installation and operation of “any particular technological system” at the source for achieving compliance with the a pplicable performance standard. Importantly, this language indicates absolutely no concern with EPA adoption of performance standards that might ever impose compliance obligations beyond the fence line of affected source. Second, section 111(h) of the CAA, which authorizes EPA to establish design, equipment, work practice, and operational standards, clearly evidences congressional intent to limit EPA’s BSER determination to inside-the-fence systems of emission reduction. For example, where EPA has determined that prescribing or enforcing a numerical performance standards would not be feasible, EPA may issue a design, equipment, work practice, or operational standard “which reflects the best technological system of continuous emission reduction.” 42 Such standards must “assure the proper operation and maintenance of any such element of design or equipment.” 43 In addition, the phrase “not feasible to prescribe or enforce a standard of performan ce” is defined as “any situation in which [EPA] determines that,” inter alia, “a pollutant or pollutants cannot be emitted through a conveyance designed and constructed to emit or capture such pollutant, or that any requirement for, or use of, such a conveyance would be inconsistent with any Federal, State, or local law.” 44 The use of the 17 phrases “technological system of continuous emission reduction,” “element of design or equipment,” and “conveyance designed and constructed to emit or capture [a] pollutant” in section 111(h) strongly implies that Congress intended EPA to limit its BSER consideration to conveyances, devices, and other such at the-unit features of design or equipment that could be used to reduce emissions. 2. Past EPA Rulemakings The proposed rule is a radical departure from past EPA rulemakings. In fact, EPA has never before issued emissions guidelines under section 111(d) that are based on the assumption that regulated sources could apply beyond -the-fence systems of emission reduction. Rather, EPA has based prior section 111(d) emissions guidelines only on measures that can be implemented at the site of the affected facility. 45 This long rulemaking history under section 111(d) confirms that BSER must be limited to at-the-unit improvements. In the preamble to this proposal, EPA notes a few of the prior section 111(d) guidelines have authorized the use of emission allowance crediting and trading to comply with guidelines. 46 However, this past rulemaking precedent does not provide support for the consideration of beyond-the-fence measures in setting the federal emissions guidelines. Rather, as noted above, none of EPA’s currently effective regulations is based on application of a beyond -the-fence BSER. 47 Indeed, the only currently effective emission guideline authorizing beyond-the-fence compliance – the emission guideline for Large Municipal Waste Combustors – is based on application of a combination of at-the-unit technological approaches and “best combustion practices,” which can be implemented only at and by the source. 48 Under this emission guideline, once 18 BSER was set based on at-the-source measures, any “state may establish a program to allow owners or operators of municipal waste combustor plants to engage in trading of nitrogen oxides emissions.” 49 Likewise, EPA’s past NSPS rules for new stationary sources under section 111(b) – which are governed by the same definition for “standard of performance” that applies to existing sources under section 111(d) – have been based exclusively on the application of at-the-unit systems of emission reduction. 50 This is most recently reflected by EPA’s proposals to regulate CO 2 from power plants under section 111. These NSPS proposals directly contradict the pos ition EPA has taken in this proposed rule. For example, EPA’s proposed NSPS for new 51 and modified 52 EGUs both consider only at-the-unit systems of emission reduction in determining BSER. EPA has not adequately explained why it has defined BSER so broadly in this proposed rule while at the same time retaining the more natural, textually supported reading of the term in other proposals to regulate CO 2 and other pollutants under section 111. This stark departure from EPA’s previous rulemaking precedent under section 111 demonstrates the unreasonableness and arbitrariness of EPA’s interpretation that forms the basis for its proposed section 111(d) guidelines based on beyondthe-fence measures. As the Supreme Court recently explained in Util. Air Regulatory Grp. v. EPA, “[w]hen an agency claims to discover in a long-extant statute an unheralded power to regulate a significant portion of the American economy, we typically greet its announcement with a measure of skepticism.” 53 Where the statutory language or EPA’s regulatory history raises legitimate questions over EPA’s authority to regulate, “[w]e expect Congress to speak clearly if it wishes to assign to an agency decisions of vast economic and 19 political significance.” 54 Even assuming that section 111(d) could be read to allow EPA to exercise the expansive, new-found authority it claims -- to regulate how the electric power system is dispatched, to mandate the amount of renewable generation to be built, and to require reduced electricity consumption -- the dictates of statutory construction demand that such significant authority be clearly established. In this case, neither the statute nor EPA’s prior interpretations of section 111 support such an expansive view of EPA’s authority. Consequently, EPA’s interpretation of section 111 which allows it to base mandatory emission guidelines on beyond-the-fence measures is illegal and should be withdrawn. 3. Removal of the Word “Technological” from Section 111’s Definition of Performance Standard EPA attempts to justify its unprecedented interpretation of section 111(d) by claiming that the addition and subsequent removal by Congress of the word “technological” from section 111’s definition of “standard of performance” indicates that EPA may consider systems other than “technological systems” when determining BSER. 55 However, even if this legislative history could be read to support the idea that BSER may include non-technological systems of emission reduction, it clearly does not support EPA’s conclusion that the CAA authorizes beyond-the-fence measures. Rather, EPA’s legal memorandum fails to consider that the removal of the term “technological” during the 1990 Amendments from the definition that applied to new sources 56 was simply meant to clarify that other available at-the-unit methods of reducing emissions other than “technological” end-of-stack improvements (such as fuel switching to lower-sulfur coals) could be considered in establishing performance standards. 20 A brief review of the legislative history of section 111 demonstrates conclusively that the narrower interpretation represents the true intent of Congress. The phrase “best system of emission reduction” was first used in the 1970 Amendments to the Clean Air Act. 57 The official Summary of the Provisions of Conference Agreement on the 1970 CAA Amendments describes section 111 as authorizing EPA to require sources to achieve performance standards “based on the latest available control technology, processes, operating methods, and other alternatives.” 58 Similarly, the 1970 Senate Committee on Public Works’ description of the Senate provisions that would later be combined into section 111 59 clearly indicates that the absence of the word “technological” from the original definition of “standard of performance” was meant to allow consideration of additional at-the-unit process and operational changes that would not have been considered “technological.” According to this Committee Report, “‘[s]tandards of performance’ . . . refers to the degree of emission control which can be achieved through process changes, operation changes, direct emission control, or other methods.” 60 Likewise, the same Committee Report makes it clear that both technological and non-technological systems of emission reduction were to be applied at existing sources: “Emission standards developed under this section would be applied to existing stationary sources. However, the Committee recognizes that certain old facilities may use equipment and processes which are not suited to the application of control technology .” 61 Because the definition of “standard of performance” in the 1970 Amendments and current law are nearly identical, and because the definition applicable to existing sources (unlike the definition for new sources) has changed only slightly since 1970, it is clear that the absence of the word “technological” from the definition of “standard of performance” in the current statute does not mean or imply that 21 beyond-the-fence measures can be designated as BSER. Rather, it is only meant to clarify that non-“technological” systems of emission reduction could also be considered BSER – so long as they could be “applied” at and by the regulated source. The addition and subsequent removal of the word “technological” to the requirements for new sources does not alter the analysis. Congress added this word to section 111 in the 1977 Amendments out of concern that power plants would lack incentives to install “technological” methods of pollution con trol (such as flue gas desulfurization systems for controlling SO 2 emissions) because they could meet the then-applicable emission limits by substituting from highsulfur coal to low-sulfur coal. 62 Congress added the word “technological” to signal to EPA that future NSPS could not be based solely on use of nontechnological, at-the-unit compliance options such as fuel switching from highsulfur fuel to low-sulfur fuel. However, the addition in 1977 and subsequent removal in 1990 of this term to the requirement for new sources did not change the fundamental intent of the statute, which was and is to limit required control techniques under section 111 to those that can be implemented at the unit and under the control of the owner or operator of the regulated source. The legislative history of the 1990 Amendments further confirms this interpretation. For example, section 501 of the 1990 Amendments clearly uses the term “technological” as a contrast to other methods of at -the-unit emission reduction approaches, such as “work practices.” 63 In addition, language in several pre-1990 Committee Reports demonstrates that Congress used the term “technological” as a way of differentiating flue gas desulfurization systems, electrostatic precipitators, and other end-of-stack treatments (which Congress 22 clearly considered to be “technological”) from other types of at -the-unit processes and systems that a regulated source could use to reduce emissions. 64 By contrast, there is no indication anywhere in the legislative histo ry that Congress intended to authorize EPA to consider beyond-the-fence systems of emission reduction. 4. EPA’s Implementing Regulations for Section 111(d) EPA’s own implementing regulations for section 111(d) – which were promulgated five years after the passage of the 1970 Amendments, also clearly contemplate that emission guidelines would be based on at -the-unit systems of emission reduction. To begin with, the regulations require EPA’s emission guideline to reflect “the application of the best system of emission reduction . . . that has been adequately demonstrated for designated facilities . . . .” 65 In addition, the regulations require EPA to provide in formation on the costs and effects of “applying each system [of emission reduction] to designated facilities” 66 as well as the time needed “for the design, installation, and startup of identified control systems.” 67 This language from EPA’s own regulations conforms with EPA’s and Congress’ past, consistent position that the term BSER can encompass only technologies and practices that can be implemented at the regulated source and under the control of the source’s owner or operator. In the context of regulating CO 2 emissions from existing EGUs, examples of such at-the-unit methods should principally focus on methods to improve the unit’s heat rate and/or lower parasitic consumption of electricity used for pollution control equipment and other ancillary facilities. Each of these methods involves the application of some technology or practice and would be within the full 23 control of the EGU. Redispatch, RE additions, and EE, by contrast, are neither at-the-unit improvements nor are they usually within the source’s control. EPA has not adequately explained why it has chosen the broadest possible interpretation of this authority, rather than a more reasonable interpretation that is in line with both the statutory text and legislative and regulatory history. B. EPA May Not Consider Building Blocks 2, 3 or 4 Because These Control Measures Are Not Part of an Inside-the-Fence System of Emission Reduction that is Within the Source’s Control. EPA proposes to set each state’s CO 2 emissions rate goal based on four emissions reduction measures, referred to as Building Blocks in the proposed rule. Three of these four Building Blocks would involve measures that are not part of an inside-the-fence system of emission reduction that is within the regulated source’s control. These measures include re-dispatch from coal-fired to NGCC generation (Building Block 2); increased use of RE (or other nonemitting generation), completion of new nuclear units, and preservation of existing nuclear (Building Block 3); and increased EE (Building Block 4). As discussed below, EPA’s reliance of these three Building Blocks in setting each state’s emissions rate goal is illegal because these control m easures are not part of an at-the-unit, inside-the fence system of emission reduction that is within the source’s control. Under the plain terms of section 111(d), the federal emissions guidelines must be based on the “best system of emission reduction;” BSER must be both “achievable” and “adequately demonstrated” for any existing source. 68 In addition, under section 111, any designated system of emission reduction must be 24 “applied” to the affected source. 69 To be achievable and adequately demonstrated for a source, a system of emission reduction must, at the very least, be capable of being implemented by the source in a “reasonably reliable, reasonably efficient,” and reasonably cost-effective way. 70 Beyond-the-fence measures that require the participation of entities other than the owner or operator of affected sources cannot meet these requirements because the implementation of those measures cannot be achieved or controlled by the source or its owner or operator. This is clearly the case with respect to the measures identified in Building Blocks 2, 3, and 4. The owner or operator of a typical coal-fired power plant cannot require consumers to use less electricity by implementing EE measures in order to reduce its own emissions. It cannot require RE developers to install additional capacity, nor require the dispatch of that RE as replacement energy for coal- or other fossil-fueled generation. It cannot directly control whether existing nuclear facilities must be retired for technological or economic reasons, decide whether new ones are built, nor force a NGCC unit to increase its generation. In addition, beyond-the-fence measures are not reliable because, first, major shifts from coal-fired generation to NGCC and RE generation will present significant challenges to reliability of the electric grid that are beyond the control of owners and operators of power plants. For example, plant owners and operators typically cannot control whether additional pi peline capacity will be constructed to enable them to increase generation at NGCC units to a level that exceeds the historic utilization of those units. In addition, the major shifts in the location and mix of generation resources that the proposed rule will require (e.g., from large, base load coal-fired units to dispersed RE or demand 25 response resources), could pose significant reliability risks and will only be possible with major investments in new transmission and distribution infrastructure. The operators and owners of affected fossil -fueled generators, however, cannot control whether or how such reliability and transmission concerns are addressed. Consequently, none of the proposed Block 2, 3, or 4 “systems of emission reduction” can be considered either “reliable” or “achievable” by “any existing source,” as required by the statute and existing case law. In addition, none of these “systems” can be “applied” by or at the affected source. In contrast to all of EPA’s previous section 111 rules, which have relied on implementation and operation of at-the-unit emission reduction systems that can be fully controlled by the regulated source, 71 the measures that EPA has identified in Building Blocks 2, 3, and 4 of the proposed BSER determination are distinctly beyond the control of the source and therefore cannot qualify as adequately d emonstrated, achievable systems of emission reduction that can be applied at regulated sources. Consequently, EPA’s reliance on these beyond -the-fence measures in its BSER determination exceeds the agency’s legal authority under section 111(d) and provides clear grounds for the invalidation of any rule based on such measures. C. Building Block 2 Cannot Form the Basis for State Emission Rate Goals Because Re-Dispatch Would Not Lead to Continuous Reductions. Even assuming that EPA could consider re-dispatch from coal-fired power plants to NGCC generation as one “system of emission reduction,” EPA may not 26 designate re-dispatch as BSER because this “system” would lead to an increase in CO 2 emissions from some affected stationary sources. Under section 111(d), states establish “standards of performance” for existing stationary sources. The CAA contains two definitions of the term “standard of performance.” Section 111(a)(1) defines a “standard of performance” as “a standard for emissions of air pollutants which reflects the degree of emission limitation achievable through the application of [BSER].” 72 Section 302(l) (a general section of definitions applicable to the entire CAA), meanwhile, defines “standard of performance” to mean “a requirement of continuous emission reduction, including any requirement relating to the operation or maintenance of a source to assure continuous emission reduction.” 73 Although the CAA includes two definitions for the term, these definitions are not in conflict. Therefore, under standard rules of statutory construction, these definitions must be read together. 74 Because these definitions must be read together, a standard of performance under section 111(d) must be based on BSER and require “continuous emission reductions.” Although the CAA does not define the term “continuous,” it should be given its ordinary meaning. The Merriam-Webster Dictionary defines “continuous” to mean “marked by uninterrupted extension in space, time, or sequence.” 75 Therefore, in order to satisfy the plain language of the CAA, state performance standards must require continuous, or constant, reductions in emissions of CO 2 from all affected sources. EPA’s proposal to base BSER on redispatch from coal to natural-gas fired units would directly contradict this requirement. By basing its guidelines on an assumption that existing NGCC units will increase their utilization through redispatch, EPA has effectively proposed to require states to establish standards of performance that would allow – indeed, require – many affected sources to 27 increase their CO 2 emissions. Thus, EPA’s proposal to incorporate redispatch into the BSER on which states must base their standards of performance is in direct conflict with the CAA’s requirement that performance standards ensure “continuous emission reduction.” D. EPA’s Alternative Consideration of Interpretation Does Beyond-the-Fence Not Measures Support in the Building Blocks 2, 3 and 4. EPA proposes two statutory interpretations in support of its position that it may set state CO 2 emission rate goals based on beyond-the-fence measures. The first is based on an interpretation that the statute generally authorizes the consideration of the control measures in Building Blocks 2, 3, and 4 because they are part of an interconnected electricity sector and result in reduced utilization, and therefore reduced emissions, from higher emitting fossil fueled -fired power plants. 76 The discussion above clearly demonstrates why this interpretation is incorrect. Most importantly, the statutory language specifically requires the emission reduction systems apply to and achieved by the affected stationary source. As a result, BSER is limited to only those measures that may be undertaken at affected power plants, such as heat rate improvements under Building Block 1, and cannot include measures that are beyond each affected plant, such as those measures in Building Blocks 2, 3, and 4. Recognizing the weaknesses inherent in this first interpretation, EPA has developed an alternative interpretation of the statute whereby BSER consists of the heat rate improvement measures in Building Block 1, plus “reduced utilization” of the affected fossil-fueled power plants at levels that are 28 “commensurate with Building Blocks 2, 3, and 4.” 77 In support of this interpretation, EPA argues that the statutory phrase “system of emission reduction” includes reduced generation because the term “system” is defined as “a set of things working together as … as an interconnecting network” 78 and that “reduced generation is a ‘set of things’ – which include reduced use of generating equipment and therefore reduced fuel input – that the affected may take to reduce its CO 2 emissions.” 79 Such an interpretation stretches the notion of a “system” of emission reduction far beyond its reasonable meaning. The simple act of turning off or retiring a stationary source of air pollution and replacing its generation with another source cannot be equated with the implementation of a “system of emission reduction” at the regulated source. In fact, ceasing operation does not even meet EPA’s own overly broad definition of a “system” that is described above. 80 Shutting down a power plant to reduce emissions does not constitute “[a] set of things working together as parts of a mechanism or interconnecting network,” EPA’s definition for a “system.” 81 Neither the statutory language nor legislative history provides any support for EPA’s alternative interpretation. Neither directs or otherwise authorizes EPA to reduce emissions “by any means necessary” or “in any way possible.” Rather, as described above in subsection (a), both the statutory language and legislative history contemplate EPA adopting regulations that require affecte d stationary sources to reduce their emissions through the installation and operation of a “system” of emission reduction at the affected source. If it is to have any meaning at all, the term “system of emission reduction” must be distinguishable in some way from the simple decision not to operate or to reduce the utilization 29 of an emitting facility. 82 EPA has not explained what, if any, limitations it believes the terms of section 111 place on its authority to order affected units to shut down or lower their production levels in order to reduce emissions. Could, for example, EPA base emission limits for petroleum refineries on an assumption that they could stop or reduce production of gasoline due to increased gasoline mileage requirements? Arguably, under the EPA alternative approach, the reduced utilization of refineries might be justifiable based on the assumption that reduced demand for gasoline could be achieved through increased gasoline mileage requirements or the increased electrification of the transportation sector. Clearly, such interpretations of section 111 would strain the notion of a “standard of performance” based on a “system of emission reduction” to the breaking point. Yet, taken to its logical conclusion, EPA’s position that BSER includes reduced utilization at EGUs could lead to similarly absurd results in other major industrial sectors subject to section 111(d) regulation. Such an interpretation is arbitrary, capricious, and otherwise contrary to the law and intent of the CAA. EPA’s alternative interpretation is fundamentally inconsistent with the main purposes of the CAA and section 111. The objective of the CAA was not to shut down or reduce utilization of existing power plants. Rather, it was to ensure that industry could continue to expand and provide economic value as it gradually reduced the amount of pollution per unit of economic output. This purpose can be seen in section 111’s mandate that EPA consider costs and energy requirements in prescribing standards of performance. It is also evident in section 111(d)’s requirement that EPA permit states to take into consideration 30 the “remaining useful life” of existing sources. Had Congress intended that EPA could simply order polluting sources to shut down or reduce production levels in order to reduce emissions, it would have had no need to require EPA to consider costs, energy, and the remaining useful life of sources in the regulation of those sources. Indeed, the existence of these requirements is a clear indication that Congress intended that section 111 would not be used to require facilities to shut down or reduce production levels in order to achieve specific levels of air pollution. In spite of this clear congressional intent to regulate air pollution in a manner that avoids unreasonable economic impacts on entire industries, EPA’s proposed rule would lead to significant stranded costs as coal -fired power plants – many of which have recently invested in expensive upgrades – are forced to retire under EPA’s guidelines. For example, EPA’s own calculations assume that eleven states will be required to shut down every coal -fired power plant in the state in order to comply with the state goal. 83 Several other states would be required to reduce coal generation by more than 5 0 percent. The state goals in the proposed rule, if adopted as EPA has formulated them, would require a reduction in annual electricity generation from existing coal units of more than 376 TWh, or of 25 percent versus the 2012 baseline. 84 A reduction of this magnitude could lead to dozens of shutdowns of plants that have just recently made multi-million dollar investments in new pollution control technology to comply with EPA’s MATS rule. Utilities that have invested in expensive emissions controls for mercury and other hazardous air pollutants will thus be faced with billions of dollars in stranded costs as a result of EPA’s proposed rule. The imposition of such significant costs on owners of coal -fired 31 power plants is neither reasonable nor appropriate in light of the goals of the CAA. In addition, under section 111(d), EPA’s regulations must allow states, in applying standards of performance to existing sources, to consider the “remaining useful life” of those sources. 85 Yet EPA’s goals, if enacted, would effectively take away states’ ability to consider the remaining useful life of the large number of coal-fired EGUs that EPA itself admits would be required to retire. Consequently, the large number of retirements that would be necessitated by EPA’s guidelines provides further evidence that EPA’s approach to determining BSER exceeds the agency’s statutory authority. Finally, EPA’s assertions in its Legal Memorandum that reduced utilization qualifies as a “system of emission reduction” under section 111 because other provisions of the CAA provide for reduced utilization or retirement as a means of compliance 86 are unconvincing. EPA has conflated the concept that allowances and retirements can be used to comply with an emission target with the concept that retirements or reduced utilization can be relied upon in determining the target in the first place. It is clear from the language of section 111(d) that states and power plants may use flexible compliance options—including the choice of purchasing emission credits or retiring—in order to comply with state plan performance standards. However, it is equally clear from t he language and history of the Act, as explained above, that EPA may not base the federal emissions guidelines on an assumption that retirements or reduced utilization are the “best system of emission reduction” under section 111. EPA has failed to provid e any convincing evidence from either the statutory language, EPA’s own rulemaking history, or the legislative history to support its proposed “alternative” BSER 32 determination. Consequently, EPA’s proposal that reduced utilization could be considered BSER is unreasonable and contrary to law. III. EPA’s Beyond-the-Fence Interpretation of the Statute Would Impermissibly Usurp States’ Traditional Sovereign Powers to Regulate the Generation, Transmission, Distribution, and Use of Electricity. Section 111 of the CAA grants EPA authority only to regulate emissions of air pollutants from stationary sources within the affected source category. These statutory powers do not, however, extend to the regulation of energy, such as the generation, transmission, distribution, and consumption of electricity within any particular state. Yet, despite the lack of any clear authorization in the CAA to regulate energy, the proposed rule would dramatically expand EPA’s power over traditional state energy functions, in violation of long-running presumptions against construing statutes to allow such broad expansions of federal regulatory authority. Long-standing Supreme Court precedent makes it absolutely clear that neither EPA nor any other federal agency may regulate the entire electricity sector without express statutory authority to do so. Under well-established principles of federalism, courts and agencies may rely on the Supremacy Clause to override state prerogatives only if Congress has been “unmistakably clear” in its grant of authority. 87 Moreover, decisions regarding electricity generation, distribution, and consumption are traditionally within the sovereign authority of states, and EPA has not historically regulated these areas. 88 As the Supreme Court stated in Pacific Gas & Electric Co., the “[n]eed for new power facilities, their economic feasibility, and rates and services, are areas th at have been 33 characteristically governed by the States.” 89 The Court further noted that “[w]ith the exception of the broad authority of … the Federal Energy Regulatory Commission, over the need for and pricing of electrical power transmitted in interstate commerce, these economic aspects of electrical generation have been regulated for many years and in great detail by the states.” 90 To be sure, there are situations in which Congress has directed a federal agency to regulate matters that are traditionally reserved to states under the tenth amendment to the Constitution. Notable examples include federal laws authorizing the conversion of existing power plants from natural gas or oil to coal 91 and prohibiting construction of new base load power plants without certain capabilities. 92 However, as noted above, well-established Supreme Court precedent clearly bars a federal agency from overriding or infringing upon a traditional state sovereign function unless Congress has adopted “unmistakably clear” statutory language that expressly authorizes the Agency to do so. 93 Neither section 111 nor other provision of the CAA contains any explicit, let alone “unmistakably clear,” authorization for EPA to regulate energy matters pertaining to the generation and use of electricity. Notably, section 111 deliberately confines EPA’s authority to the areas in which its expertise is greatest: the evaluation of technologies to reduce or eliminate emissions; the applicability of those technologies to various classes and categorie s of sources; the cost of employing such technologies; and the energy and non -air environmental impacts associated with the use of those technologies. This lack of express authority further confirms Congress’ intent to respect and preserve the states’ historic role over the regulation of electricity and thereby not to 34 authorize EPA to adopt a CO 2 performance standard that effectively dictates how electric utilities generate, transmit and distribute electricity for end use. In conclusion, nowhere does section 111 permit EPA to issue regulations that would rely on actions taken by state regulators, RE providers, end-use consumers, or other entities that EPA asserts comprise potential pathways to compliance with the EPA guidelines. Indeed, even assuming arguendo that the statute is ambiguous enough to allow EPA to take such an expansive approach to its power over stationary source emissions, the presumption against federal intrusion onto a traditional state power requires that EPA proffer a clear statement to support its authority. The CAA supplies no such clear statement, 94 and EPA’s assertion of authority is therefore impermissible in light of state responsibilities over electricity generation, distribution, and consumption wh ich the proposed rule, if finalized, would displace. IV. EPA Has Not Met the Legal Prerequisites to Regulate Existing Power Plants Under Section 111(d) of the Clean Air Act. A. EPA Does Not Have Authority to Regulate CO 2 Emissions from Existing Power Plants Because it Has Failed to Make the Proper Endangerment Finding for the EGU Source Category. Section 111 clearly states that EPA has no authority to regulate any source category of stationary sources until it makes a finding that the particular source category “causes, or contributes significantly to, air pollution which may reasonably be anticipated to endanger public health or welfare.” 95 However, the proposed rule has failed to meet the CAA’s requirements for making such a 35 finding for the electric generating unit (EGU) source category for greenhouse gas emissions. First, the proposed rule ignores a key statutory distinction between the endangerment finding requirement for regulating the mobile source category under section 202(a) of the CAA and the endangerment finding required for regulating the EGU source category under section 111 of the Act. Second, the proposed rule’s attempt to rely on the mobile source endangerment finding fundamentally misconstrues EPA’s own determination regarding greenhouse gases. Since an endangerment finding specific to the EGU source category is a necessary prerequisite for regulation of greenhouse gas emissions from power plants, both of these errors provide further grounds for invalidating any final rule that EPA may adopt to regulate existing power plants under section 111(d). 1. EPA Has Ignored a Key Statutory Distinction Between the Endangerment Findings that Are Required for Mobile and EGU Source Categories. EPA issued an endangerment finding for greenhouse gas emissions from mobile sources in 2009. 96 The proposed rule now attempts to rely on this 2009 endangerment finding by making general reference to the health and welfare impacts that EPA cited in its 2009 finding. 97 However, an endangerment finding for mobile sources under section 202(a) does not satisfy the endangerment finding requirement for power plants under section 111 because the relevant statutory language of the provisions is not identical. Section 202(a) of the CAA requires the Administrator to prescribe standards for pollutants from classes of motor vehicles which “cause, or contribute to, air pollution which may reasonably be anticipated to endanger public health or welfare.” 98 In contrast, 36 section 111(b)(1)(A) instructs the Administrator to include a category of sources which “causes, or contributes significantly to, air pollution which may reasonably be anticipated to endanger public health or welfare.” 99 There are well-established rules of statutory interpretation that require agencies and courts to construe the text of a statute in a manner that avoids "surplusage” and gives “full and obvious meaning” to each and every word of the statute. 100 Here, EPA has effectively tried to read the word “significant” out of the statute by blurring the mobile- and stationary-source endangerment standards together. Such an interpretation would do violence to the language of the statute by conflating two distinct standards and thereby ignoring the clear statutory standard that the cause or contribution to endangerment must be “significant” under section 111. Furthermore, this interpretation runs contrary to the recent decision by the Supreme Court in EME Homer City Generation, in which the Supreme Court made it clear that EPA must carefully consider the statutory role of the term “significant” when interpreting its authorizing statutes. 101 While significance is “not a mathematical straitjacket,” and the agency has discretion in determining which risks are acceptable and which are not, EPA has a legal obligation to make an actual finding that a significant contribution from the regulated sources is present before regulating under section 111(d). 102 In the proposed rule, EPA has failed make such a finding. The proposed rule contains no evaluation of the relationship between atmospheric CO 2 levels – which are determined on a global level – and the contribution of the EGU source category that EPA proposes to regulate. Absent a demonstration by EP A that that relationship is “significant,” the endangerment finding cannot be valid under 37 the plain language of the statute and prior court interpretations of the term “significance.” Finally, it is also noteworthy that the CAA establishes other regulat ory programs or requirements that are predicated on separate endangerment findings. For example, a state’s obligation to develop state implementation plans under section 110 of the CAA to attain national ambient air quality standards (NAAQS) only begins upon EPA making an endangerment finding and then promulgating a NAAQS for a pollutant under sections 108 and 109 of the CAA. 103 Similarly, an endangerment finding is also required as prerequisite for the regulation of fuel additives, non-road engines and non-road vehicles, and aircraft engines under the CAA. 104 To be sure, the Act does not contain a single endangerment protocol that can then be applied in each of these regulatory contexts; rather, the statutory language clearly demonstrates that Congress established different endangerment requirements throughout the statute. To remain consistent with the statutory structure of distinct endangerment findings across a range of CAA regulatory programs and requirements, EPA must treat power plants as an independent category, and not merely seek to bootstrap one endangerment finding onto another. 2. The Proposed Rule Misapplies the 2009 Endangerment Finding for Mobile Sources. Another fatal flaw is that EPA has fundamentally misconstrued its prior mobilesource endangerment finding in its attempt to rely on that finding in making an endangerment finding for CO 2 emissions from power plants under the proposed rule. As a result, EPA impermissibly builds its regulations on an incongruent 38 endangerment finding that cannot support the proposed rule. While EPA argues in the preamble to the proposal only that CO 2 emissions from fossil-fueled power plants endanger public health and welfare, the 2009 endangerment finding did not specifically determine this. Instead, the very first sentence of the 2009 endangerment finding states that with respect to mobile sources, “six greenhouse gases taken in combination endanger both the public health and the public welfare of current and future generations.” 105 Further, the endangerment finding repeats several times that “six well-mixed greenhouse gases” pose health and welfare risks. 106 Even in the proposed rule, EPA reviews the endangerment to health and welfare caused by greenhouse gases generally, not CO 2 in particular. 107 EPA’s first proposed CO 2 rule in 2012 for new EGUs at least referenced in passing the fact that the 2009 endangerment finding addressed six GHGs, 108 and the re-proposed January 2014 rule for new EGUs responded briefly to comments regarding the inadequacy of EPA’s endangerment finding under the earlier proposal. 109 However, sometime between January 2014 and June 2014, all reference to “six well-mixed” greenhouse gases simply vanished. Although EPA cites the 2009 endangerment finding extensively in the proposed rule, EPA appears to ignore the actual scope of that finding at every opportunity. This is an impermissible leap under the law. Even if the statutory requirements for endangerment findings under sections 202(a) and 111(b) were identical – which, as noted above, they are not – EPA’s actions in the proposed rule are entirely incongruent with its prior finding. The 2009 mobile source endangerment finding did not determine that each of the six identified greenhouse gases individually threaten health and welfare. Rather, EPA’s use of the term “well mixed” was deliberate and consistent throughout the finding, and EPA may not 39 simply ignore the specific elements of its endangerment finding today because it finds them inconvenient. B. CAA Prohibits EPA from Regulating Existing Power Plants Under Section 111(d) if There Is No Valid, Effective New Unit Rule for Power Plants in Place. Section 111(d) expressly provides that EPA may only regulate an existing source “for any air pollutant” if the source is one “to which a standard of performance under [section 111] would apply if such existing source were a new source. . . .” 110 Thus, under the plain terms of the statute, EPA may not regulate CO 2 from existing power plants unless it has established new source performance standards (NSPS) for CO 2 for new power plants. regulations implementing section 111(d) confirm this EPA’s own interpretation. 111 Therefore, if EPA does not finalize its proposed CO 2 NSPS for new power plants, 112 or if a court subsequently vacates this CO 2 NSPS rule following judicial review, EPA will by operation of law lack authority to issue or enforce CO 2 emissions guidelines for existing plants under section 111(d). In the proposed rule, EPA asserts that even if the CO 2 NSPS for new power plants were not finalized or vacated, EPA could still move forward with section 111(d) regulation of existing power plants so long as it has finalized CO 2 performance standards for modified power plants. 113 However, EPA’s position that performance standards for modified plants could substitute for new unit standards contradicts the plain language as well as the intent of the s tatute. As discussed, the plain text of section 111(d) provides that only existing sources that would be subject to performance standards if they were new sources can be 40 regulated. The statute does not state that sources that would be subject to performance standards if they were modified sources may be regulated. Moreover, it would be unreasonable to interpret the statute to authorize EPA to regulate only modified and existing power plants without first regulating new power plants. The Senate Committee that drafted the original text of section 111(d) stated at the time that the purpose of NSPS is to “provide maximum incentives to expand technology to insure adequate margins of safety,” and to require “[t]he maximum use of available means of preventing and controlling air pollution” when capital investments are being made to construct new sources. 114 Were EPA to regulate existing and modified sources before regulating new sources, the Agency could effectively shirk its responsibility to set the NSPS benchmark for state-of-the-art pollution controls for new sources – a result that would be inconsistent with the basic structure of section 111. Therefore, EPA’s conclusion that it may regulate existing EGUs if new EGUs are not subject to CO 2 performance standards is unreasonable and contrary to both the text and intent of the statute. Finally, we note that if EPA’s NSPS for new power plants is invalidated upon judicial review, the invalidation would also immediately invalidate any final section 111(d) rule that the Agency may have previously adopted for existing plants. Under section 111(d), EPA may only require states to regulate existing sources for which performance standards would “apply if such existing source were a new source.” 115 Without a final CO 2 NSPS remaining in legal effect for new power plants, this statutory prerequisite would no longer be satisfied because there would no longer be a legally effective CO 2 standard applying to new plants. Therefore, if EPA’s proposed rules for new power plant s are 41 subsequently invalidated, EPA’s guidelines for existing power plants would also cease to have legal effect. V. The CPP will have Significant and Unacceptable Economic Consequences. Cost is one key factor that EPA must consider in setting the emission targets under its BSER analysis. 116 Furthermore, the courts have made it clear that the cost of the “control system” under the BSER analysis may not be “exorbitantly costly” and that any costs that are “unreasonable” or “excessive” would indicate that the control system in question is not the “best.” 117 As discussed below, the proposed CPP would impose massive costs on the electric power sector and American consumers in violation of this clear statutory limitation placed on EPA’s authority to regulate existing power plants under section 111(d) of the CAA. According to analysis conducted by NERA Economic Consulting, the proposed CPP will be the most expensive environmental regulation imposed on the electric sector in history. ACCCE, along with the American Fuel & Petrochemical Manufacturers, Association of American Railroads, American Farm Bureau Federation, Electric Reliability Coordinating Council, Consumer Energy Alliance and National Mining Association, retained NERA to analyze the potential impacts of the CPP using their NewERA model. 118 NERA projects, in the primary scenario, that the total compliance costs of the CPP will be $366 billion between 2017 and 2031, with an annual average cost of $41 billion between 2020 and 2030. 119 The results of NERA’s analysis are outlined briefly below, while a more detailed discussion of the methodology and results of 42 NERA’s analysis is presented in the NERA report in Appendix 2 to these comments. NERA analyzed two main scenarios. The first, called the “State Unconstrained” scenario, allowed states to select the least-cost compliance option based on all four Building Blocks proposed by EPA. That scenario corresponds most closely to the scenario analyzed by EPA using IPM. The second scenario, called the “State Constrained” scenario, required that states meet EPA’s proposed targets using only Building Blocks 1 and 2. This scenario reflects the impacts of state compliance plans that do not rely on RE or EE policies to meet EPA’s proposed state targets. States, for example, may be unable or unwilling to impose federally enforceable control measures that require substantial increases in RE generation or reductions in electricity use through implementation of EE measures. 120 In addition, legitimate legal questions have been raised as to whether EPA has the authority to require reductions under Building Blocks 3 and 4 under a federal plan in the event that a state fails to submit an approvable plan under the CPP. NERA’s modeling analysis primarily relied on Energy Information Administration (EIA) Annual Energy Outlook (AEO) 2014 forecasts and costs, and EPA assumptions in all but a few instances. 121 Based on this modeling analysis, NERA projected that the CPP will impose significant costs on electricity customers. In the case in which states could use all four Building Blocks, NERA estimates a present value cost of $366 billion over the period of 2017 - 2031. 122 The annual average cost over the period of 2020 – 2030 is $41 billion. These costs are driven primarily by $560 billion in cumulative EE costs, half of which, $280 billion, will be directly borne by individual residential, 43 commercial and industrial electricity consumers. In the second scenario, in which states cannot consider additional RE or EE, the costs are even higher. NERA estimates the present value costs could be $479 billion from 2017 – 2031, or $73 billion per year over the 2020-2030 period. 123 The costs in the second scenario are driven by increases in electricity prices and natural gas price s. Table 1 – Energy System Cost Impacts of State Compliance Scenarios (billion 2013 dollars) The CPP will cause a significant increase in electricity prices. In the State Unconstrained scenario, NERA is projecting that the CPP will cause electricity prices to be 12 percent higher each year compared to the reference case. Individual states face even greater impacts. For example, 43 states will face average double digit percentage electricity price increases between 2020 and 2030. In addition, 14 states will experience peak year increases in electricity prices exceeding 20 percent over the same period. Average and peak year electricity prices between 2020 and 2030 can be fo und in Table 2 below. 44 Table 2 – Average and Peak Year Electricity Price Increases by State in the State Unconstrained Scenario Average Peak Year Average Peak Year Alaska 14% 20% Montana 19% 20% Alabama 15% 19% North Carolina 11% 13% Arkansas 17% 20% North Dakota 14% 17% Arizona 14% 15% Nebraska 15% 19% California 12% 12% New Hampshire 4% 15% Colorado 17% 19% New Jersey 11% 19% Connecticut 6% 16% New Mexico 16% 19% DC 9% 13% Nevada 18% 22% Delaware 10% 19% New York 3% 19% Florida 16% 18% Ohio 10% 18% Georgia 15% 20% Oklahoma 18% 21% Hawaii 5% 8% Oregon 19% 20% Iowa 16% 23% Pennsylvania 13% 21% Idaho 18% 18% Rhode Island 4% 16% Illinois 13% 25% South Carolina 11% 13% Indiana 14% 15% South Dakota 14% 17% Kansas 12% 18% Tennessee 14% 18% Kentucky 14% 17% Texas 10% 17% Louisiana 16% 20% Utah 24% 26% Massachusetts 5% 17% Virginia 13% 16% Maryland 10% 18% Vermont 13% 13% Maine 3% 18% Washington 16% 17% Michigan 12% 13% Wisconsin 16% 17% Minnesota 15% 18% West Virginia 12% 14% Missouri 14% 20% Wyoming 22% 26% Mississippi 14% 16% 45 National average electricity prices are 17 percent above the baseline between 2017 and 2031 in the State Constrained scenario. Between 2020 and 2030, 21 states have average electricity price increases over 20 percent, with 25 states experiencing peak year electricity price impacts above 20 percent. Average and peak year electricity prices between 2020 and 2030 can be found in Table 3 below. 46 Table 3 – Average and Peak Year Electricity Price Increases by State in the State Constrained Scenario Average Peak Year Average Peak Year Alaska 8% 8% Montana 24% 26% Alabama 16% 18% North Carolina 8% 9% Arkansas 23% 24% North Dakota 27% 30% Arizona 13% 14% Nebraska 26% 29% California 11% 12% New Hampshire 5% 5% Colorado 27% 28% New Jersey 16% 18% Connecticut 4% 5% New Mexico 16% 18% DC 6% 8% Nevada 32% 33% Delaware 20% 23% New York 9% 22% Florida 20% 25% Ohio 8% 10% Georgia 16% 18% Oklahoma 22% 25% Hawaii 8% 8% Oregon 10% 12% Iowa 27% 30% Pennsylvania 19% 31% Idaho 12% 13% Rhode Island 5% 5% Illinois 26% 30% South Carolina 8% 9% Indiana 10% 15% South Dakota 26% 29% Kansas 25% 29% Tennessee 17% 19% Kentucky 26% 33% Texas 63% 95% Louisiana 22% 24% Utah 46% 55% Massachusetts 4% 5% Virginia 13% 15% Maryland 6% 8% Vermont 9% 9% Maine 5% 6% Washington 9% 10% Michigan 20% 22% Wisconsin 18% 20% Minnesota 24% 27% West Virginia 18% 21% Missouri 23% 26% Wyoming 42% 49% Mississippi 18% 19% 47 The CPP will also lead to significant increases in coal retirements. NERA projects 45,000 MW of incremental coal retirements due to the CPP under the State Unconstrained scenario. The State Constrained scenario results in 169,000 MW of incremental coal retirements. By 2030 a total of between 96,000 and 220,000 MW of coal capacity may retire according to NERA’s analysis. These additional coal retirements will threaten electric reliability, discussed below. VI. The Proposed State Emission Targets are Based on Arbitrary and Technically Incorrect Building Blocks and These Errors Provide Further Compelling Grounds for EPA Withdrawal of the Proposed Rule. As discussed above, EPA lacks authority to regulate existing power plants under section 111(d) of the CAA. EPA also may not consider beyond-the-fence measures in determining BSER and setting the state emission targets. Although these flaws alone require the withdrawal of the CPP, t his section discusses additional technical and legal problems with each of the Building Blocks that EPA has proposed to use in setting state CO 2 emissions targets. These problems include faulty Building Block assumptions and major inconsistencies in the application of those assumptions in setting the goals among the states. As a result, in addition to being illegal from a statutory perspective, EPA’s proposed approach to setting state goals is also arbitrary, capricious, and unreasonable and therefore provides further compelling grounds for EPA withdrawal of the proposed rule. 48 A. EPA has Incorrectly Estimated the Potential Heat Rate Improvement Available at Existing Coal-Fired Power Plants Under Building Block 1. In setting CO 2 emissions rate targets for states, EPA incorrectly assumed all existing coal-fired power plants could achieve a six percent improvement in heat rate under Building Block 1. 124 The assumed six percent heat rate improvement target is based on two factors. First, EPA presumed power plants can achieve a four percent heat rate improvement by adopting “best practices” based on EPA’s statistical analysis of heat rate variability. 125 Second, EPA presumed an additional two percent heat rate improvement based on a 2009 Sargent & Lundy study of the heat rate impacts of various coal-fired power plant equipment upgrades. EPA further uses 16 “reference units” to support its claim that a six percent heat rate improvement is available at all coal-fired power plants. EPA has incorrectly interpreted this information as supporting the availability of a fleet-wide six percent heat rate improvement at existing coal-fired power plants. In contrast to EPA’s assertion, a six percent heat rate improvement is unrealistic and unachievable for most coal-fired plants and, to the extent that some heat rate improvements may be available, the costs of implementing these upgrades at affected units would generally be prohibitive and could trigger onerous New Source Review permitting requirements as discussed below. 49 1. EPA’s Assumptions About Operating “Best Practices” are Incorrect. EPA conducted a statistical analysis to conclude that a four percent heat rate improvement is available at existing coal fired power plants if those plants were to employ “best practices.” EPA’s analysis is significantly flawed and does not support the conclusion drawn by EPA. First, EPA ignores a number of facto rs that lead to heat rate variability at power plants, stating that variability other than due to capacity factor and ambient temperature implies available “best practices” improvements. Second EPA makes overly simplistic assumptions concerning the fleet-wide potential to improve efficiency through a series of undefined “best practices.” EPA ignored several important factors that impact coal-fired power plant heat rate variability in their statistical analysis. When analyzing heat rate variability, EPA attributed 26 percent of variability to variations in load and ambient air temperature. 126 EPA erroneously assumed the remaining 74 represented the potential to improve heat rate. 127 By only focusing on load and ambient temperature, EPA ignored a series of factors that can further explain heat rate variability and which are not controllable through “best practices.” For example, the type of coal used can significant ly impact heat rate through a variety of means, including flue gas moisture content and the varying fouling and slagging potential among different coal types. 128 Additionally, EPA ignored various boiler design parameters, most importantly whether the boiler operates at subcritical or supercritical steam conditions. 129 Some supercritical boilers, for example, will be less affected by variations in load. 130 Beyond ambient air temperature, ambient water temperature, not considered by EPA, plays a 50 significant role in the heat rate of coal-fired power plants, especially those using once-through cooling systems. 131 Finally, EPA does not fully capture the capacity factor, and the related amount of cycling a coal -fired power plant may undergo. 132 In order to arrive at a conclusion that four percent heat rate improvements can be achieved through best practices, EPA made overly simplistic assumptions concerning the ability of like-situated plants to reduce heat input for every megawatt hour generated. 133 EPA divided the existing coal fleet into a variety of bins based on ambient temperature and load and presumed that each power plant in each bin could improve their performance by an arbitrary 30 percent relative to the best performing plants in that bin. 134 EPA provides no justification for such an assumption. In fact, units already have an incentive to operate as efficiently as possible. Because EPA’s assumption concerning best practices is not supported, it leads to an arbitrary conclusion that coal fired power plants can improve heat rate by four percent. 2. The Sargent & Lundy Study Does Not Show a Two Percent Heat Rate Improvement is Feasible. EPA relies on an incorrect interpretation of a 2009 Sargent & Lundy engineering analysis of the potential for heat rate improvements at existing coal units. 135 Citing the Sargent & Lundy study, EPA concludes that a two -percent heat rate improvement can be achieved by all existing coal units. The Sargent & Lundy analysis lists a series of actions that can be taken by plant owners or ope rators to improve their units’ heat rate by a certain amount and at a certain cost. 136 In order for EPA to conclude a two-percent improvement is possible, an analysis of 51 the extent to which these heat rate improvements apply to every unit in the entire existing coal fleet would be needed. Sargent & Lundy’s study does not provide this type of analysis. In fact, nowhere in the study do the authors attempt to conclude what types of improvements could be undertaken at specific units in the coal fleet, or that two percent would be representative of such an improvement. Rather, the Sargent & Lundy report uses a literature survey and vendor data to present a “laundry list” of potential or hypothetical improvements that might be achievable, as opposed to actual case-by-case analysis. In fact, contrary to EPA’s assumptions, many of the potential heat rate improvements identified in the Sargent & Lundy report are non-additive, temporal, and highly site-specific. As Sargent & Lundy explained in a recent letter to the Natural Rural Electric Cooperative Association (NRECA): Combinations of strategies to achieve heat rate improvements do not always provide heat rate improvement reductions equal to the sum of each individual strategy's heat rate improvement because many of the technologies affect, or are dependent upon, plant operating variables that are inter related. Therefore, case-by-case analyses should be conducted to determine whether the incremental heat rate improvement through the application of multiple strategies is economically justified. 137 In addition, the Sargent & Lundy study was completed in 2009, three years before the 2012 baseline year on which EPA has based its proposed state goals. 52 Thus, the improvements cited in the study—to the extent they are feasible or cost-effective at all— could have been implemented already by power plants because power plant owners have an incentive to make their units as efficient as possible. In other words, some of these improvements could already be reflected in the 2012 baseline, which means that few additional heat rate improvements will be achievable in practice. Based on these serious methodological shortcomings, EPA erred in concluding the Sargent & Lundy study supports a two-percent heat rate improvement as a part of Building Block 1. 3. EPA Incorrectly Interpreted Heat Rate Data from the 16 “Reference Units.” EPA incorrectly relied on 16 “reference units” to support their claims that a six percent heat rate improvement is achievable at all existing coal-fueled power plants. After reviewing the reference units, EPA concludes that the units have already achieved a three to eight percent heat rate improvement. 138 However, of the 16 reference units selected by EPA, an analysis of 14 of those units suggests that EPA’s incorrectly interpreted the heat rate data from those units. 139 Ten of the fourteen plants analyzed report that the heat rate improvements observed by EPA were not reflective of actual improvements in heat rate performance achieved by the plants, but instead represented variability in, or problems with, the accuracy of the Continuous Emissions Monitors (CEMS) from which heat rate improvement data was derived. 140 In addition, in some cases in which EPA determined heat rate improvement had occurred, improvements were later 53 offset by installation of pollution control equipment required by subsequent regulations, like CSAPR and MATS. 141 4. EPA’s Building Block One Analysis Fails to Consider the Impacts of Other Building Blocks. Yet another problem with EPA’s Building Block 1 assumption is that it fails to consider the impact of implementing the other Building Blocks under the proposed rule. In particular, EPA has failed to address the interaction of Building Block 1’s heat rate improvements with Building Block 2’s re-dispatch to natural gas and Building Block 3’s increased deployment of renewable and nuclear energy. The most significant heat rate improvements naturally occur when units are able to operate as close as possible to their rated capacity at steady state; 142 however, the implementation of Building Blocks 2, 3, and 4 would decrease the average capacity factor of coal-fired steam EGUs, forcing them to run at lower levels. Furthermore, the increased use of RE will clearly result in coal-fired generation cycling up and down in order to balance intermittent loads from new wind and solar generation. The need to run at lower capacity factors, when combined with increased cycling of units, will greatly decrease the efficiency of these same units for which EPA is assuming a six percent heat rate improvements under Building Block 1. 5. EPA has Ignored the Barrier that NSR Poses to Efficiency Improvements at Existing Coal-Fired Power Plants. In Utility Air Regulatory Group v. EPA, the Supreme Court held that EPA could require a “best available control technology” (BACT) analysis for greenhouse 54 gas emissions for sources, including power plants, that trigger such a review for other air pollutants. 143 A BACT evaluation is required for sources that trigger NSR for conventional air pollutants in areas that are in attainment, or for pollutants with no ambient air quality standards, like CO 2 . In defining BACT, the Act specifically requires that BACT be no less stringent than standards for such a facility under section 111 or 112 of the Act. 144 As a result, power plants that trigger NSR for conventional pollutants may be subject to the proposed greenhouse gas emission standards for modified and reconstructed coal plants even though they would not otherwise qualify for such standards. Because of the relationship to NSR, the proposed greenhouse gas emissions standards for modified coal-fired power plants would serve as a strong disincentive for power plants to undertake efficiency improvements. The threat of NSR and being required to meet those proposed standards will discourage power plants from undertaking any meaningful efficiency improvements. EPA has previously noted the negative impact of NSR on efficiency impro vements. 145 Notably, the disincentive provided by NSR is largest for those plants that are least efficient because compliance at those plants would be the most expensive. EPA ignores that fact that those proposals increase the disincentive to improve efficiency at existing coal-fueled power plants, in contrast to the efficiency improvements projected in Building Block 1. EPA incorrectly analyzed historical heat rate data, misinterpreted an engineering analysis, and erroneously attributed heat rate improvement at 16 reference units to conclude that a six percent heat rate improvement was available at all existing coal-fueled power plants. Additionally, EPA has failed to factor in the adverse impacts of the other Building Blocks on the heat rates of 55 existing coal-fired units in estimating the potential for efficiency improvements at those units. As a result of these errors, EPA’s conclusion that a six percent heat rate improvement is available at all coal-fired power plants is incorrect. B. EPA Incorrectly Assumed that Existing NGCC Power Plants Can Operate at 70 Percent Capacity Factors under Building Block 2. EPA has incorrectly assumed that all existing NGCC power plants can achieve, on average nationwide, an annual capacity factor of 70 percent under Building Block 2. 146 EPA has either failed to consider, or incorrectly considered, a number of factors that limit the ability of existing NGCC to reach a 70 percent capacity factor on an annual basis. 1. There is No Basis for Assuming All Existing NGCC Units Can Achieve a 70 Percent Capacity Factor Due to Historical Data, Market Dynamics and Reliability Concerns. There is no basis to support the assumption that all existing NGCC units could reach, on average, a 70 percent capacity factor. Except for a small number of units, most existing NGCC units do not operate at levels that approach a 70 percent capacity factor. In fact, when examining regional NGCC capacity factors since 2005, capacity factors range from 12.6 percent to 63.4 percent. 147 Even during 2012, when natural gas prices were at historic lows, the U.S. fleet of NGCC units operated at an average capacity factor of less than 50%. 56 Table 4 – Performance Characteristics of Natural Gas Combined Cycle Units by Region: Capacity Factor % 148 New England Middle Atlantic 2005 48.0% 24.4% 2006 49.7% 27.3% Weighted Average Capacity Factor 2007 2008 2009 2010 2011 50.8% 48.2% 48.2% 55.1% 57.4% 33.9% 34.0% 42.6% 46.0% 51.2% 2012 52.8% 59.7% 2013 45.0% 52.0% East North Central 17.5% 16.2% 20.1% 14.3% 16.4% 22.0% 28.5% 48.0% 33.1% West North Central 20.4% 16.5% 25.0% 20.2% 12.6% 17.6% 14.7% 26.5% 21.5% South Atlantic w/o Florida 20.3% 21.1% 26.6% 23.8% 36.1% 33.9% 41.0% 53.7% 57.9% Florida 52.6% 57.8% 52.6% 55.0% 53.0% 59.1% 60.7% 63.4% 56.8% South Atlantic 38.4% 41.5% 41.6% 41.9% 46.5% 48.4% 52.2% 59.0% 57.3% East South Central 31.0% 36.2% 30.7% 28.0% 38.1% 43.8% 49.7% 59.3% 48.0% West South Central w/o ERCOT 28.0% 32.6% 32.9% 33.3% 36.0% 36.2% 37.9% 47.6% 35.8% ERCOT 51.8% 51.6% 52.0% 49.9% 46.5% 45.1% 47.3% 50.1% 48.0% West South Central 41.8% 43.5% 43.8% 42.8% 42.1% 41.4% 43.4% 49.2% 43.3% Mountain 40.0% 42.7% 48.2% 48.0% 45.7% 40.9% 34.2% 40.4% 37.0% Pacific Contiguous w/o CA 51.5% 44.1% 48.8% 49.7% 53.1% 51.1% 26.3% 32.9% 47.2% California 45.3% 53.4% 61.5% 61.5% 52.4% 52.7% 39.7% 55.1% 48.9% Pacific Contiguous 47.0% 51.1% 58.3% 58.3% 52.6% 52.3% 36.2% 49.6% 48.5% 34.3% 36.1% 39.2% 37.3% 38.3% 40.8% 40.8% 49.4% 42.8% Census Region TOTAL U.S. NGCC plants operate at lower capacity factors than coal-fired units because they typically serve as “intermediate load” plants. EPA significantly overestimates the potential for increased capacity factors at NGCC units when setting state CO 2 emissions rate targets. The dynamics of wholesale power markets make it difficult for existing NGCC power plants to achieve and sustain an average 70 percent capacity factor. Increased output from existing NGCCs will signal a market need for new NGCCs to provide intermediate generating capacity. These cheaper, more efficient new units will arise to replace generation from existing units before existing units can achieve an aggregate 70 percent capacity factor. 149 This dynamic is partly responsible for the historic capacity factors of Table 4, and is also in evidence in EPA’s own IPM model results. In its “Option 1 Regional” 57 results, EPA shows NGCC units will operate at a capacity factor between 55 percent (in 2018) and 59 percent (2050). 150 EPA’s “Option 1 State” results show NGCC units running between 55 percent (2018) and 60 percent (2050). 151 Another problem with EPA’s assumption that NGCC units could increas e their capacity factors to 70 percent is that doing so could compromise electric reliability on a local or regional basis. 152 NGCC units provide an important balancing function that allows utilities to incorporate large amounts of variable renewable resources. Unless utilities and system managers reserve a sufficient amount of spare NGCC capacity, it could be difficult to keep the grid operating during periods of high electric demand and low RE production . If NGCC utilization were to increase to the levels that the EPA assumes, the system could become even more vulnerable to periodic blackouts or curtailments during high demand, low-variable-supply scenarios. 153 In addition, increased reliance on natural gas could lead to periodic weather- or natural disaster-related reliability issues similar to the prolonged shutdown of natural gas facilities in the East, Midwest, and Southwest during severe cold snaps in previous winters. 154 2. EPA Ignores Engineering and Infrastructure Issues that Prevent Increased NGCC Generation, and Includes Erroneous Data About NGCC Capacity. Many NGCC units were not constructed for base load operations and could require significant evaluation and upgrades to enable them to operate continuously at a 70 percent or greater capacity factor on an annual basis. 155 Even if existing NGCC units are capable of running at 70 percent or greater capacity factors, gas pipeline capacity may not be sufficient to provide the gas 58 volumes needed to operate at these high capacity factors, especially in the short term before additional infrastructure can be built. This lack of available natural gas infrastructure will limit the availability of natural gas for all consumers, including NGCC power plants. This will make it difficult for NGCC power plants to achieve the capacity factors EPA assumes. Gas supply infrastructure constraints may be further exacerbated during extreme weather events such as the past winter’s Polar Vortex, when larger than normal gas supplies needed for space heating reduced the gas available for power generation. 156 Furthermore, EPA assumes that this dramatic increase in the dispatch of NGCC generation is fully achieved at the beginning of the interim compliance period in 2020. This major shift in generation to NGCC over the next five years is infeasible for many existing units because of technical limitations, gas pipeline infrastructure constraints, and/or re gulatory limits (discussed below). Given these constraints on increasing the capacity of existing NGCC units, EPA’s proposed timeline to implement Building Block 2 by the start of the interim compliance period in 2020 is infeasible. The significant shi ft EPA contemplates from coal to natural gas would require significant investment in gas pipeline infrastructure to accommodate the increased demand for natural gas. Gas pipeline construction and upgrades require permits and approvals at the federal, state, and local levels that can take five years or more to secure and put in place. The timeframe that EPA has allowed for this Building Block does not recognize the extended period of design, permitting, and construction that would be required to increase the utilization of existing NGCCs. It is unlikely that new infrastructure would be available to support the ramp up of natural 59 gas fired units by 2020, thereby putting pressure on states and utilities to achieve more drastic emission reductions in later years of the interim compliance period to comply with the interim average emissions goal. 157 In addition, EPA made a series of errors in calculating the potential for redispatch of NGCC under Building Block 2 to reduce emissions rates in each state. For example, EPA used nameplate capacity, and not net summer capacity to predict increases in NGCC generation. 158 Such an assumption serves to overestimate the potential for additional NGCC generation since net summer capacity is lower than nameplate in most cases and is a more accurate representation of the available generation capacity of an NGCC plant. 159 Additionally, EPA included a number of NGCC plants that were not available in 2012. 160 Further, some of the plants used by EPA in calculating NGCC potential were simple cycle combustion turbines. 161 Finally, EPA ignored the constraints that grid operators and or state public service commissions may impose on the increased dispatch of NGCC plants by assuming all could reach a 70 percent capacity factor. 162 3. EPA Ignores the Fact that Environmental Permits Limit the Operation of NGCC Capacity. Operational constraints that would limit the capacity factor of existing NGCC units may exist in air permits that were written to cap emissions of NO x or carbon monoxide in ozone nonattainment areas. Source owners with emissions limitations on an NGCC unit would be prohibited from increasing capacity to the levels EPA proposes as that would potentially violate air permits. 163 In addition, increasing NGCC capacity factors could lead to conflicts as unit60 specific run-time limitations that are contained in many NGCC operating permits. Units with permit-level run-time restrictions will likely be unable to increase utilization to levels approaching 70 percent. 164 4. EPA’s Alternative Approaches for Calculating Building Block 2 are Unworkable and Illegal. In the October 2014 NODA, EPA proposes two changes to its method of calculating emission reductions achievable through redispatch to natural gas. 165 First, EPA proposes to make Building Block 2 more stringent by assuming that coal-fired power plants can implement some amount of co-firing or rely on construction of new NGCC capacity to reduce emissions from coal-fired power plants. 166 This approach is unworkable and illegal for at least three reasons. First, designating natural gas co-firing as BSER would effectively “redefine” the source by requiring coal-fired EGUs to use significant amounts of natural gas to generate electricity. Under well-established CAA precedent, EPA may not make assumptions about available control technology that would result in a “redefinition” of the source. 167 However, the reliance on natural gas co-firing in setting Building Block 2 would violate this precedent by requiring a facility designed as a coal-fired power plant to make major design and infrastructure changes in order to co-fire with natural gas. EPA cannot simply redefine an entire source category in this manner in an attempt to reduce emissions from that category. Second, EPA cannot designate new units as a component of the best system of emission reduction when setting the performance levels for existing units. To 61 do so would impermissibly blur the clear statutory distinction b etween new and existing units. That distinction makes clear that under section 111, an “existing source” is “any stationary source other than a new source.” 168 Third, as EPA itself recognizes, building new NGCC or co -firing is expensive. EPA estimates the cost of reducing emissions through co-firing to be as high as $150 per metric ton of CO 2 . 169 The cost of reducing emissions through redispatch to new NGCC units, meanwhile, would be as high as $81 per metric ton. 170 Therefore, it would be unreasonable to designate these costly alternatives as the “best system of emission reduction (considering the cost of such reduction) that has been adequately demonstrated for designated facilities. . . .” 171 EPA also requests comment on whether it should quantify the level of NGCC availability on a regional, rather than state-by-state basis. 172 Although this approach might better resemble the way that power markets work than EPA’s current proposal, ACCCE cannot reasonably evaluate this proposal unless EPA provides additional details about how it would develop the regions, how it would model power transfers and transmission capacity among states, how it will address the fact that different ownership structures and contractual arrangements for NGCC facilities will affect each unit’s ability to reduce the need for other generation elsewhere, and other questions. Until EPA provides the public with a better understanding of what it is proposing, ACCCE cannot comment on this issue. In conclusion, there is no basis for assuming existing NGCC can operate at a 70 percent capacity factor. In addition, EPA has ignored infrastructure constraints and permit limitations. Similarly, EPA’s alternative approach is unworkable 62 and illegal. All of these factors indicate EPA incorrectly assumed existing NGCC power plants can operate at 70 percent capacity factor in Building Block 2. C. EPA Incorrectly Assessed the Potential for States to Increase RE as a Part of Building Block 3. EPA has developed two approaches for the RE target for each state under Building Block 3. 173 Under the first approach, EPA divides the country into regions and sets the target level for all states within a region based upon an average of the 2020 Renewable Portfolio Standards (RPS) from the states within each region that have enacted an RPS. Under second approach, the agency sets the RE target by determining both the technical and market potential for RE within each state. For the reasons discussed below, the methodology used in setting RE targets under each approach is fundamentally flawed and is therefore not a reliable indicator of the potential for RE generation in each state. 1. There Are Major Methodological Flaws in Setting the RE Targets for States Under the First Approach. Under the first approach, EPA’s assessment of the potential for RE to contribute to CO 2 emissions targets is based on several incorrect assumptions. approach is arbitrary and unreasonable for at least three reasons. EPA’s First, EPA ignored the details of state RPS that would limit the overall RE potential calculated by EPA. Second, EPA arbitrarily applied regional averages based on state RPS to states without renewable targets, ignoring the different resource 63 potentials in each state. Third, EPA ignores costs and other technical limitations that reduce the potential for RE growth. EPA ignored the variability in state RPS and therefore overestimated the national potential for RE growth in Building Block 3. State legislatures have allowed different types of generation to contribute to meeting state RPS. For example, hydroelectricity contributes more than 18 percent of New York’s 29 percent RPS. 174 Additionally, many states preferentially allow for certain types of in-state RE to receive credit for more than the amount of electricity actually generate, up to 3.5 times. 175 In some states, like North Carolina, EE programs are allowed to count toward the state RPS. 176 Further, many states also allow out-ofstate RE to be counted toward RPS compliance. 177 All of these factors would limit the actual amount of RE in an individual state, which is what EPA attempts to measure in Building Block 3. By failing to appropriately adjust its regional targets downward for these different elements of state RPS, EPA overestimates the potential for increases in RE to contribute to CO 2 emissions rate targets. EPA has arbitrarily used a regional approach to create estimates of RE potential in states that do not currently have an RPS. In states that do not have an RPS, EPA uses an average of the RPS of the states in the region to calculate the potential for RE growth. In two regions, EPA sets regional RE targets based on the RPS of only one state in those regions. 178 significant variability that exists state-to-state. In doing so, EPA ignores the 179 For example, in the South Central region, EPA ignores the fact that Kansas has significant wind resources available in applying the Kansas RPS level to Nebraska, Oklahoma, Arkansas, Texas and Louisiana. 180 While Texas and Oklahoma have wind resources similar 64 to those in Kansas, Louisiana and Arkansas have very low potential wind resources. 181 Similarly, EPA arbitrarily applies the RPS in North Carolina to the entire Southeastern region of the United States. Instead of undertaking any analysis of the potential for expanded RE in states, EPA arbitrarily applies a regional average of state RPS programs to states that do not currently have an RPS target. FIGURE 1 – 2020 Effective State RPS Targets Finally, EPA failed to consider any technical or economic consideration which would limit the potential growth of RE in states. For example, EPA did not conduct a renewable resource assessment for any state. 182 Similarly, EPA did not consider the infrastructure constraints that would inhibit significant increases in RE. One such limitation is the need for additional transmission capacity needed to incorporate new intermittent resources. 183 65 EPA also failed to assess the practical limitations associated with the construction of RE facilities. 184 As a result of these flaws in EPA’s analysis, EPA set RE targets for each state that are arbitrary and unreasonable estimations of renewable energy resources available in each state. 2. There Are Major Methodological Flaws in Setting the RE Targets for States Under the Alternative Approach. EPA has proposed an alternative approach for setting each state’s RE target, which is based on both technical potential and market potential for RE within a state. Under this second approach, EPA determined technical potential by scaling each state’s theoretical technical potential for each type of RE generation, based on a study by the National Renewable Energy Laboratory, to the average development rate for that technology among the top 16 states nationwide. 185 As EPA concedes, this “technical potential” is alone insufficient to demonstrate that any particular level of RE development is adequately demonstrated because data about theoretical technical potential ignores restrictions such as “grid limitations, costs associated with development, and quality of resource.” 186 The “market potential” for deployment of each RE technology represents the “market potential for RE generation that can be developed in each state,” not what can be economically deployed and dispatched. 187 EPA has attempted to estimate the amount of generation that a state could deploy in light of cost constraints, but EPA’s approach to developing the market potential figures is seriously flawed and therefore cannot be used by EPA in setting the RE target under the final rule. 188 66 Notably, this alternative RE approach results in RE goals for a number of states that are in excess of 40 percent (excluding existing hydropower) of the states’ 2012 retail sales. 189 In two states, Kansas and South Dakota, the RE goal exceeds 100 percent (excluding existing hydro) of 2012 retail sales. This extraordinarily high level of intermittent RE has no precedent in actual electric system operations and therefore does not appear to be remotely feasible. Reliability impacts, infrastructure requirements, and the question of whether there is even a market for the projected amount of electricity must be taken into account before setting a state’s goal. The excessively high levels of these RE goals clearly indicate the unworkability of the EPA alternative approach for setting RE goals for states. EPA failed to properly assess the variability in state RPS programs, arbitrarily applied a regional RPS level to states that do not have a program , and ignored all practical considerations that would limit potential RE growth. Additionally, EPA’s proposed alternative approach contains significant methodological flaws. As a result, EPA acted arbitrarily when it considered the potential for RE growth in Building Block 3. D. EPA Incorrectly Assessed the EE Potential as a Part of Building Block 4. EPA’s assessment of the potential for EE to contribute to state CO 2 emissions rate targets in Building Block 4 is significantly overestimated, and therefore incorrect. 190 First EPA has incorrectly concluded that the studies it reference s support a 1.5 percent year-on-year increase in EE. Second EPA failed to account for the EE programs that are already in the baseline. Finally, EPA makes no 67 distinction between the potential for EE to be deployed in the commercial, industrial and residential sectors. The studies on which EPA relies to set the contribution of EE programs to state CO 2 emissions rate targets do not support EPA’s conclusion. EPA merely averages results from twelve studies to conclude that 1.5 percent year-on-year growth in EE is achievable, resulting in demand reductions of around 10 percent in every state. 191 reasons. This approach is arbitrary and unreasonable for several For example, the studies upon which EPA relies contain multiple scenarios. EPA does not attempt to rectify the different scenarios in each study, but merely averages across all the scenarios without consideration of the assumptions therein. 192 In so doing, EPA fails to adjust for impacts to the potential EE growth rate under different assumptions. EPA also ignores the fact that these studies have different base years , some of which are in the past. 193 The selection of base year is an important consideration because many subsequent actions to improve the efficiency across sectors ha ve already been taken; therefore the EE growth that would be projected by the studies with earlier base years overestimate the total potential for EE growth. 194 In addition, EPA does not adjust for the time period the studies cover, which will alter how returns on EE programs are evaluated. 195 EPA has also not appropriately considered the extent to which existing state and federal EE programs have already been considered in EIA’s baseline forecasts. EIA implicitly incorporates EE into their baseline electricity demand forecasts. 196 EIA’s consideration of EE can be evidenced through their measurement of energy intensity, typically measured in consumption per dollar of GDP. 197 EIA 68 has attributed about one quarter of the overall improvement in energy intensity to these programs. 198 For example, many of the gains achievable by improving lighting efficiency have already been used. 199 In addition, some states have eliminated their EE programs because the targets set by the state were unachievable. 200 By failing to adjust for the EE growth that has already occurred and is already in the baseline, EPA significantly overestimates the potential growth of EE in Building Block 4. EPA also failed to address the different potential EE adoption available to different types of electricity consumers. Many of the studies cited by EPA break down the potential for increases in EE by residential, industrial and commercial sectors. 201 In doing so, those studies recognize the different potential for improving efficiency that exists in the different sectors. 202 Generally speaking, the residential sector has the highest potential for improving efficiency, while industrial and commercial consumers have lower potential to improve. 203 Even among the residential consumers, potential varies significantly based on the region due to the weather and use of natural gas for home heating and certain appliances. 204 The industrial and commercial sectors already have a profit motive for improving efficiency and have therefore already adopted many energy efficiency measures. Therefore, the states in which industrial or commercial electricity consumers make up a large portion of electricity consumption will have limited EE improvements available. 205 By failing to differentiate potential EE growth by sector in each state, EPA arbitrarily ignored the important differences in opportunity that exist to improve EE in different settings. 69 EPA incorrectly concluded that the twelve studies it considered support a 1.5 percent year-on-year increase in EE, failed to consider EE measures that have already occurred, and ignored the different potential for EE that exists among different sectors. As a result, EPA’s assumptions about the potential for EE in Building Block 4 are incorrect. E. EPA Applied the Four Building Blocks Arbitrarily to Determine Each State’s Target Emission Rate. The methodology used by EPA to calculate target emission rates for each state under the CPP relies on historical operating data from 2012. These data include existing generating units and their output levels, as well as power demand and interstate power transfers. However, the electric power system is dynamic, and its operation in one time period has little bearing on how it will operate in another time period. Factors such as fuel prices for power plants, demand variability, and the construction or retirement of electric infrastructure will all affect the generation levels for individual power plants. Selecting a single year as a starting point from which to calculate target emission rates will “lock in” any transient effects particular to that year, regardless of the year chosen. In its October 2014 NODA, EPA released power plant operating data for years 2010 and 2011. Although EPA did not release sufficient data to recreate state emission rate targets using these years as the base, the data showed a substantial difference in power plant operations from those of 2012. Certainly, some states will find their target rates higher or lower if a year other than 2012 is used as a base year. Importantly, EPA’s methodology includes adjustments for power imports and exports between states; for consistency, this requires the 70 same base year to be applied to every state, and would not permit a state to choose its own base year without leading to inconsistent target emission rates in other states. Therefore, the arbitrary and capricious nature of selecting a single year to determine emission rate targets cannot be mitigated without a fundamental change to EPA’s methodology. VII. The CPP Will Have a Negative Impact on Electric Reliability. Several organizations that are responsible for electric reliability, or grid operation, have noted that the CPP is expected to pose significant threats to electric reliability. The North American Electric Reliability Corporation (NERC) has released an initial review of the Clean Power Plan and concludes that it poses reliability issues. Similarly the Southwest Power Pool (SPP), the Midcontinent Independent System Operator, the Electric Reliability Corporation of Texas (ERCOT) and American Electric Power (AEP) have released initial assessments indicating the CPP will result in reliability threats. NERC, the organization which develops and enforces reliability standards for the bulk power system in the United States and Canada, has concluded that the CPP will pose adverse risks to electric reliability in several areas. 206 NERC points to significant coal-retirements and a corresponding shift to natural gas and RE as posing threats to reliability. This change can cause issues with reserve margins, vulnerability to natural gas supply constraints, and the need for increased essential reliability services. 207 NERC also notes that the significant change to the electric system that could be brought about by the CPP will require additional investment in transmission infrastructure, which typically takes 5 to 15 years to complete, much longer than the few years i n 71 which a state plan is developed and compliance begins. NERC also concluded that EPA incorrectly assessed the potential emissions reductions achievable by each of the four Building Blocks. 208 SPP, MISO and ERCOT, organizations that operate wholesale electricity markets within and across multiple states, have also highlighted the potential negative impacts the CPP will have on electric reliability. SPP has estimated that CPP will result in reserve margins of 4.7 percent by 2020 and -4.0 percent by 2024, well below their statutory requirements of 13.7 percent. 209 Similarly, MISO predicts that the significant emissions reductions needed by the beginning of the compliance period will result in challenges to resource adequacy under the CPP. 210 ERCOT highlights the potential reliability impacts of coal-fired power plant retirements and increased RE generation on the ERCOT grid. 211 Additionally, AEP, a utility that operates in eleven states, notes that the CPP could lead to thermal overloads, voltage collapse and othe r reliability impacts. 212 All five of these organizations, responsible for maintaining reliability standards for the grid, operating the grid, or generating electricity, have raised significant concerns about the impact the CPP will have on electric reliabi lity. VIII. The CPP Will Have No Meaningful Impact on Climate Change. Despite its significant costs, the CPP will have no meaningful impact on climate change. In order to understand the impact the CPP will have on climate change, ACCCE applied the cumulative reduction of CO 2 , to the impact estimates used by EPA in the 2012 emissions standard for light duty vehicles. 213 When using EPA’s own methodology, the impacts are meaningless. For example: 72  Atmospheric CO 2 concentrations would be reduced by 1.52 parts per million (ppm), or less than one-half of 1 percent, in 2050. For perspective, the IPCC projects concentrations of 450 ppm to 600 ppm in 2050.  Global average temperature increase would be reduced by less than two hundredths of a degree (0.016° F) in 2050. The IPCC projects an increase of 1.8° to 3.6° F in 2050.  Sea level rise would be reduced by 0.3 millimeter (the thickness of three sheets of paper), or 1/100th of an inch. The IPCC projects a sea level rise of 5.9 inches to 11.8 inches in 2050. 214 Moreover, when EPA attempts to quantify the benefits of its proposal, EPA significantly overstates the value of reducing emission s by using the Social Cost of Carbon (SCC). ACCCE submitted comments in February 2014 on the flawed approach taken by EPA, and other federal agencies, in attempting to estimate the SCC. 215 One of the most significant errors in the SCC estimate is the use of global benefits instead of domestic benefits. Should EPA use a domestic SCC, the overall benefits would be 7 percent to 23 percent of the values used by EPA. 216 For the CPP, the climate benefit estimated by EPA of $31 billion would be reduced to $2 billion to $7 billion, far below the $41 to $73 billion estimated annual cost projected by NERA. 217 Conclusion The proposed Clean Power Plan should be withdrawn for a number of reasons. EPA has no legal authority to regulate coal-fired power plants under section 111(d), and cannot consider reductions that could be achieved “outside -thefence.” All four of the Building Blocks EPA used to calculate state CO 2 73 emissions rate targets are based on arbitrary and fundamentally flawed assumptions and tend to overstate the potential emissions reductions available. Moreover, the CPP is the most expensive environmental regulation ever issued for the electricity industry, will pose significant threats to electric reliability, and is of no consequence to global climate change. For t hese legal and policy reasons, we respectfully request EPA that withdraw the proposed Clean Power Plan. 1 79 Fed. Reg. 34,830 (June 18, 2014). 2 A list of ACCCE members is provided in Appendix 1. 3 Carbon Pollution Emission Guidelines for Existing Stationary Sources: Electric Utility Generating Units; Proposed Rule, 79 Fed. Reg. 34,830 (June 18, 2014). J. Edward Cichanowcz, Critique of EPA’s Use of Reference Units to Select Heat Rate Reduction 4 Targets (October 13, 2014). 5 Energy Ventures Analysis, Redispatching Gas Combined Cycle Units (November 2014). 6 Energy Ventures Analysis, Review of EPA Building Block #3A – Renewable Energy (December 2014) Energy Ventures Analysis, Assessment of EPA Building Block 4 – Energy Efficiency (December 7 2014) In 2010, EPA promulgated the Mercury and Air Toxics Standards, then the most expensive 8 environmental regulation for the electric power sector at $10 billion per year. EPA, Regulatory Impact Analysis for the Final Mercury and Air Toxics Standards (December 2011) ($9.6 billion compliance cost in 2006$ converted to 2010$.) In 2007, EPA estimated that all Clean Air Act regulations for the electric power sector cost $7 billion. EPA, The Benefits and Costs of the Clean Air Act from 1990 to 2020 (2011) at Table 3-2 (electric utility direct annual compliance costs were $6.6 billion (2006$) in 2010; this converts to $7.1 billion in 2010$). NERA, Potential Energy Impacts of the EPA Proposed Clean Power Plan (October 2014). 9 10 North American Electric Reliability Corporation, Potential Reliability Impacts of EPA’s Proposed Clean Power Plan (November 2014); Southwest Power Pool, SPP’s Reliability Impact Assessment of the EPA’s Proposed Clean Power Plan (October 8, 2014); Midcontinent Independent System 74 Operator, Analysis of EPA’s Proposal to Reduce CO 2 Emissions from Existing Units (November 12, 2014); ERCOT, ERCOT Analysis of the Impacts of the Clean Power Plan (November 17, 2014); American Electric Power, Transmission Challenges with the Clean Power Plan (September 2014). 11 ACCCE, Climate Effects of EPA’s Proposed Carbon Regulations , June 2014. 12 42 U.S.C. § 7411(d)(1) (2012). 13 See Chevron U.S.A., Inc. v. Natural Res. Defense Council, Inc. , 467 U.S. 837 (1984). 14 See 104 Stat. 2468, Pub. L. No. 101-549. 15 42 U.S.C. § 7411(d)(1). 16 See Carbon Pollution Emission Guidelines for Existing Stationary Sources: Electric Utility Generating Units, Proposed Rule, 79 Fed. Reg. 34,830, 34,853 (June 18, 2014) [hereinafter “Proposed Rule”]. 17 Revision of December 2000 Regulatory Finding on the Emissions of Hazardous Air Pollutants From Electric Utility Steam Generating Units and the Removal of Coal - and Oil-Fired Electric Utility Steam Generating Units from the Section 112(c) List, F inal Rule, 70 Fed. Reg. 15,994, 16,031 (Mar. 29, 2005) [hereinafter “CAMR Rulemaking”], vacated by New Jersey v. EPA, 574 F.3d 574 (D.C. Cir. 2008) on other grounds, cert. denied, 129 S. Ct. 1313 (2009). 18 See, e.g., Revisor’s Note, 5 U.S.C. app. 3 § 12; Revisor’s Note, 7 U.S.C. § 2018; Revisor’s Note, 8 U.S.C. § 1324b; Revisor’s Note, 10 U.S.C. § 869; Revisor’s Note, 10 U.S.C. § 1074a; Revisor’s Note, 10 U.S.C. § 1407; Revisor’s Note, 10 U.S.C. § 2306a; Revisor’s Note, 10 U.S.C. § 2533b; Revisor’s Note, 11 U.S.C. § 101; Revisor’s Note, 12 U.S.C. § 1787; Revisor’s Note, 12 U.S.C. § 4520; Revisor’s Note, 14 U.S.C. ch. 17 Front Matter; Revisor’s Note, 15 U.S.C. § 1060; Revisor’s Note, 15 U.S.C. § 2081; Revisor’s Note, 16 U.S.C. § 230f; Revisor’s Note, 18 U.S.C. § 1956; Revisor’s Note, 18 U.S.C. § 2327; Revisor’s Note, 20 U.S.C. § 1226c; Revisor’s Note, 20 U.S.C. § 1232; Revisor’s Note, 20 U.S.C. 4014; Revisor’s Note, 21 U.S.C. § 355; Revisor’s Note, 22 U.S.C. § 2577; Revisor’s Note, 22 U.S.C. § 3651; Revisor’s Note, 22 U.S.C. § 3723; Revisor’s Note, 23 U.S.C. § 104; Revisor’s Note, 26 U.S.C. § 105; Revisor’s Note, 26 U.S.C. § 219; Revisor’s Note, 26 U.S.C. § 613A; Revisor’s Note, 26 U.S.C. § 1201; Revisor’s Note, 26 U.S.C. § 4973; Revisor’s Note, 26 U.S.C. § 6427; Revisor’s Note, 29 U.S.C. § 1053; Revisor’s Note, 33 U.S.C. § 2736; Revisor’s Note, 37 U.S.C. § 414; Revisor’s Note, 38 U.S.C. § 3015; Revisor’s Note, 39 U.S.C. § 410; Revisor’s Note, 40 U.S.C. § 11501; Revisor’s Note, 42 U.S.C. § 218; Revisor’s Note, 4 2 75 U.S.C. § 300ff-28; Revisor’s Note, 42 U.S.C. § 3025; Revisor’s Note, 42 U.S.C. § 5776; Revisor’s Note, 49 U.S.C. § 47115. 19 See, e.g., Am. Petroleum Inst. v. SEC, 714 F.3d 1329, 1336-37 (D.C. Cir. 2013) (declining to infer contrary congressional intent from an apparent clerical error in renumbering subsections of the Securities Exchange Act through subsequent amendment). 20 To Amend the Clean Air Act to Provide for Attainment and Maintenance of Health (Enrolled Bill) , S.B. 1630 § 302(a), available at: http://thomas.loc.gov/cgi- bin/query/D?c101:5:./temp/~c101B3dcEU:: 21 U.S. Senate, Office of the Legislative Counsel, Legislative Drafting Manual § 126(b)(2)(A) (Feb. 1997), http://www.law.yale.edu/documents/pdf/Faculty/SenateOfficeoftheLegislativeCounsel_Legis la tiveDraftingManual(1997).pdf. 22 42 U.S.C. § 7412(c). 23 See id. 24 42 U.S.C. § 7412(n)(1)(A). 25 CAMR Rulemaking, 70 Fed. Reg. at 16,031. 26 See id. 27 Am. Bar Ass’n v. FTC, 430 F.3d 457, 469 (D.C. Cir. 2005). 28 EPA, Legal Memorandum for Proposed Carbon Pollution Emission Guidelines for Existing Electric Utility Generating Units 23 (June 2014), http://www2.epa.gov/sites/production/files/2014 06/documents/20140602-legal-memorandum.pdf [hereinafter “Legal Memorandum”]. 29 Reiter v. Sonotone Corp., 442 U.S. 330, 339 (1979) (when construing statutes, courts and executive agencies must “give effect, if possible, to every word Congress used”); see also Carcieri v. Salazar, 555 U.S. 379, 391 (2009) (same); United States v. Menasche, 348 U.S. 528, 53839 (1955) (same). 30 Cf. Am. Bar Ass’n v. FTC, 430 F.3d 457, 469 (D.C. Cir. 2005). 31 Scialabba v. Cuellar de Osorio, ___ U.S. ___, 134 S. Ct. 2191, 2218 (2014) (Sotomayor, Breyer, & Thomas (partial), JJ., dissenting (citing A. Scalia & B. Garner, Reading Law: The Interpretation of Legal Texts 180 (2012)); see also United Savings Ass’n v. Timbers of Inwood Forest Associates , 484 U.S. 365, 371 (1988) (“Statutory construction . . . is a holistic endeavor. A provision that may seem ambiguous in isolation is often clarified by the remainder of the statutory scheme — 76 because the same terminology is used elsewhere in a context that makes its meaning clear, or because only one of the permissible meanings produces a substantive effect that is compatible with the rest of the law.”) (citations omitted); United States v. Boisdoré’s Heirs, 49 U.S. 113, 122 (1850) (“In expounding a statute, we must not be guided by a single sentence or member of a sentence, but look to the provisions of the whole law, and to its object and policy.”). 32 Scialabba, 134 S. Ct. at 2218 (quoting FTC v. Mandel Bros., Inc., 359 U.S. 385, 389 (1959)); id. at 2214 (Roberts & Scalia, JJ., concurring) (quoting FDA v. Brown& Williamson Tobacco Corp., 528 U.S. 120, 133 (2000)). 33 Scialabba, 134 S. Ct. at 2214 (Roberts, & Scalia, JJ. concurring); id. at 2220 n.3 ( Sotomayor & Breyer, JJ. dissenting); id. at 2216 (Alito, J. dissenting). 34 Am. Bar Ass’n v. FTC, 430 F.3d 457, 469 (D.C. Cir. 2005). 35 Util. Air. Regulatory Grp. v. EPA, ___ U.S. ___, 134 S. Ct. 2427, 2444 (2014) (quoting Brown & Williamson Tobacco, 529 U.S. at 159)). 36 Id. (quoting Brown & Williamson, 529 U. S. at 160) (internal quotation marks omitted). 37 Id. 38 See, e.g., Legal Memorandum at 23. As we explain above, in our view, the 1990 CAA Amendments clearly prohibit EPA from using section 111(d) to regulate pollutants emitted from source categories regulated under section 112. 39 42 U.S.C. § 7411(d)(1)(A) (emphasis added). 40 42 U.S.C. § 7411(a)(1) (emphasis added). 41 42 U.S.C. § 7411(b)(5) (emphasis added). 42 42 U.S.C. § 7411(h)(1). 43 Id. (emphasis added). 44 42 U.S.C. § 7411(h)(2) (emphasis added). 45 See, e.g., Standards of Performance for New Stationary Sources an d Emission Guidelines for Existing Sources: Large Municipal Waste Combustors, Proposed Rule, 70 Fed. Reg. 75,348, 75,351 (Dec. 19, 2005). 46 79 Fed. Reg. 34,927. 47 Even in the vacated CAMR rule, which would have authorized states to establish a cap -and- trade system among EGUs, EPA did not base the emission guideline on beyond -the-fenceline measures of reducing emissions in the electric sector such as renewable energy additions or 77 energy efficiency. Rather, the state mercury “budgets” in CAMR were calculated based on EPA’s assumption that existing at-the-unit, end-of-stack technologies would be used to reduce mercury emissions. See generally Standards of Performance for New and Existing Stationary Sources: Electric Utility Steam Generating Units; Final Rule, 70 Fed. Reg. 28,606, 28,617-28,621 (May 18, 2005) (discussing the BSER basis for establishing the national cap); id. at 28,621 (state budgets “were developed by totaling unit-level emissions reductions requirements for coal-fired electricity generating devices.”) (emphasis added). 48 See Standards of Performance for New Stationary Sources and Emission Guidelines for Existing Sources: Large Municipal Waste Combustors, Proposed Rule, 70 Fed. Reg. 75,348, 75,351 (Dec. 19, 2005) (emission guidelines developed based on application of “spray dryer/electrostatic precipitator/activated carbon injection/selective non -catalytic reduction technology (SD/ESP/ACI/SNCR) or spray dryer/fabric filter/activated carbon injection/selective non-catalytic reduction technology (SD/FF/ACI/SNCR).”) 49 40 C.F.R. 60.33b(d)(2). 50 See, e.g., National Emission Standards for Hazardous Air Pollutants From Coal and Oil -Fired Electric Utility Steam Generating Units and Standards of Performance for Fossil -Fuel-Fired Electric Utility, Industrial-Commercial-Institutional, and Small Industrial-Commercial- Institutional Steam Generating Units; Proposed Rule, 76 Fed. Reg. 24,976, 25,060 -61 (May 3, 2011); Standards of Performance for Stationary Combustion Turbines; Proposed Rule, 70 Fed. Reg. 8,314, 8,318-20 (Feb. 18, 2005). 51 Standards of Performance for Greenhouse Gas Emissions From New Stationary Sources: Electric Utility Generating Units, Proposed Rule, 79 Fed. Reg. 1430 (Jan. 8, 2014) [hereinafter “New Unit Proposal”]. 52 Carbon Pollution Standards for Modified and Reconstructed Stationary Sources: Electric Utility Generating Units, Proposed Rule, 79 Fed. Reg. 34,960 (June 18, 2014) [hereinafter “Modified Unit Proposal”]. 53 Util. Air. Regulatory Grp. v. EPA, ___ U.S. ___, 134 S. Ct. 2427, 2444 (2014) (quoting Brown & Williamson Tobacco, 529 U.S. at 159) (citation and quotation marks omitted). 54 Util. Air. Regulatory Grp. v. EPA, ___ U.S. ___, 134 S. Ct. 2427, 2444 (2014) (citations and quotation marks omitted). 78 55 See EPA, Legal Memorandum for Proposed Carbon Pollution Emission Guidelines for Existing Electric Utility Generating Units 56 (June 2014), http://www2.epa.gov/sites/production/files/2014 -06/documents/20140602-legalmemorandum.pdf [hereinafter “Legal Memorandum”]. 56 As EPA correctly points out, the definition of standard of performance applicable to existing sources has never included the term “technological.” See Legal Memorandum at 55-56. Consequently, Congress’ decision to remove the term from the definition that previously applied to new sources has no bearing on the interpretation of the term as it applies to existing sources. 57 Clean Air Amendments of 1970, Pub L. No. 91-604, § 111, 84 Stat. 1676, 1683 (codified as amended at 42 U.S.C. § 1857 et seq. (currently codified at 42 U.S.C. § 7401 et seq.)). 58 Clean Air Act Amendments of 1970 – Conference Report, 116 Cong. Rec. 42,381, 42,384 (1970) (emphasis added). 59 60 These provisions were included in sections 113 and 114 of the Senate bill. S. Rep. No. 91-1196, at 17 (Sept. 17, 1970) (emphasis added). See also id. (“The Secretary would be directed to review and promulgate new or modified standards whenever new technology processes or operating methods become available.”) (emphasis added); H. R. 17255, 91st Cong. § 113(b)(2) (as amended and passed by Senate Sept. 22, 1970) (requiring the establishment of performance standards reflecting “the greatest degree of emi ssion control which the Secretary determines to be achievable through application of the latest available control technology, processes, operating methods, or other alternatives. ”) (emphasis added). 61 S. Rep. No. 91-1196, at 19 (emphasis added). 62 See, e.g., H.R. Rep. No. 95-294, at 188 (May 12, 1977) (“In order to effectuate the purposes stated above in the promulgation (and revision) of new source standards of performance, the committee has specified that standards under section 111(b) must require use of the best technological system which has been adequately demonstrated. This means that a major new stationary source may no longer meet NSPS requirements merely by use of untreated oil or coal. Instead, employment of a technological system would become a mandatory aspect of the NSPS requirements for such a source.”). 63 See 1990 Amendments, Pub. L. No. 101-549, § 507, 104 Stat. 2399, 2645-46 (codified as amended at 42 U.S.C. § 7401 et seq.) (requiring EPA to approve state “small business stationary 79 source technical and environmental compliance assistance” programs if the state includes, among other things, “[p]rocedures for consideration of requests from a small business stationary source for modification of . . . (A) any work practice or technological method of compliance ….”) (emphasis added). 64 See, e.g., S. Rep. No. 100-231, at 232 (Nov. 20, 1987) (“The technologies, practices or strategies which are to be considered in setting emission standards under [section 112] go beyond the traditional end-of-the-stack treatment or abatement system. The Administrator is to give priority to technologies or strategies which reduce the amount of pollution generated through process changes or the substitution of less hazardous substances. A second class of control requirements focusing on the enclosure of emissions points or other secondary containment systems is also to be considered before systems to collect, capture or treat the emissions are reviewed. Considering municipal waste combustion as a potential category of major sources to illustrate these concepts (even though standards for [Municipal Waste Combustors] are set independently by another provision of this legislation): source separation and good combustion are examples of process modifications; covered faci lities for ash handling and negative air flow are examples of enclosure and secondary containment; and scrubbers, precipitators and selective catalytic reduction are examples of end -of-the-stack treatment systems.”); S. Rep. No. 101-228, at 168 (Dec. 20, 1989) (substantially the same language). 65 40 C.F.R. § 60.22(b)(5) (emphasis added). 66 40 C.F.R. § 60.22(b)(3) (emphasis added). 67 40 C.F.R. § 60.22(b)(4) (emphasis added). 68 See 42 U.S.C. § 7411(a)(1) (standards of performance must be based on “degree of e mission limitation achievable through the application of [BSER].”); Id. § 7411(d)(1)(A) (states establish “standards of performance for any existing source ….”) (emphasis added). 69 See 42 U.S.C. § 7411(a)(1)(defining standard of performance as “a standard … which reflects the degree of emission limitation achievable through the application of [BSER].”) (emphasis added). 70 Essex Chem. Corp. v. Ruckelshaus, 486 F.2d 427, 433 (D.C. Cir. 1973) (emphasis added) . 71 See, e.g., Standards of Performance for New Stationary Sources and Emission Guidelines for Existing Sources: Large Municipal Waste Combustors, Proposed Rule, 70 Fed. Reg. 75,348, 80 75,351 (Dec. 19, 2005) (finalizing emission guidelines based on applicati on of emissions control technology at a source, not reductions that could be achieved elsewhere). 72 42 U.S.C. § 7411(a)(1). 73 42 U.S.C. § 7602(l). 74 See Reiter v. Sonotone Corp., 442 U.S. 330, 339 (1979) (courts and agencies must “give effect, if possible, to every word Congress used”); Brown & Williamson Tobacco Corp., 529 U.S. 120, 133 (2000) (citation omitted) (statutes should be interpreted to “fit, if possible, all parts into an harmonious whole”). See also Nat’l Cable & Telecomms. Ass’n v. Gulf Powe r Co., 534 U.S. 327, 335 (2002) (“specific statutory language should control more general language when there is a conflict between the two.”) (emphasis added). 75 Merriam-Webster Dictionary, http://www.merriam-webster.com/dictionary/continuous (last visited Aug. 11, 2014). 76 Legal Memorandum at page 54. 77 Legal Memorandum at pages 80-81. 78 Legal Memorandum at page 50. 79 Legal Memorandum at page 82. 80 See Proposed Rule at 34,885 (defining “system” as “[a] set of things working together as parts of a mechanism or interconnecting network.”) (citing Oxford Dictionary of English (3d ed., 2010). 81 Id. 82 In many cases, the owner or operator of a power plant cannot unilaterally decide to reduce its own utilization. For example, individual plants that participate in an RTO or ISO electricity markets that operate on an economic dispatch model often cannot control when or how much they must operate to support customer demand. Similarly, weather, a ccident, and other unexpected events can lead to situations in which a power plant must operate far more than expected pursuant to “reliability-must-run” requirements or other emergency orders that can be issued by NERC, FERC, or regional reliability entit ies. 83 See EPA Goal Computation Spreadsheet 1 (coal units in Alaska, Arizona, California, Connecticut, Massachusetts, Mississippi, Nevada, New Hampshire, New Jersey, Oregon, and Washington would generate zero MWh under EPA’s proposed state targets). 84 Id. 81 85 42 U.S.C. § 7411(d)(1)(B). 86 See Legal Memorandum at 83-85. 87 See Will v. Mich. Dep’t of State Police, 491 U.S. 58, 65 (1989) (quoting Atascadero State Hosp. v. Scanlon, 473 U.S. 234, 242 (1985)) (it is an “ordinary rule of statutory construction that if Congress intents to alter the ‘usual constitutional balance between the States and the Federal Government,’ it must make its intention to do so ‘unmistakably clear in the language of the statute.”); Id. (quoting Rice v. Santa Fe Elevator Corp., 331 U.S. 218, 230 (1947)) (“Congress should make its intention ‘clear and manifest’ if it intends to preempt the historic powers of the States.”). See also Medtronic, Inc. v. Lohr, 518 U.S. 470, 485 (1996); G ade v. Nat’l Solid Wastes Mgmt. Ass’n, 505 U.S. 88, 111 (1992). 88 Interstate transmission of electricity and wholesale sales of electricity are federally regulated by the Federal Energy Regulatory Commission, not EPA. 89 Pac. Gas & Elec. Co. v. State Energy Res. Conservation & Dev. Comm’n, 461 U.S. 190, 205 (1983). The limited grant of authority to FERC supports the notion that, should Congress intend to displace traditional state authority over the electricity sector, it would only convey that power clearly. In contrast, Section 201 of the Federal Power Act limits FERC’s jurisdiction to transmission and wholesale energy sales; FERC cannot directly regulate power generation. Similarly, Section 202 of the FPA references “voluntary interconnection and coordin ation” of electric facilities across regions; FERC cannot compel any particular sort of coordination. Atlantic City, 295 F.3d 1. If Congress wished to extend federal authority over the broader electricity system, it could have done so at the time of prom ulgation of Part II the FPA in 1935, or during any of that Act’s subsequent revisions. Instead, the primary Federal statute governing electricity clearly limits the scope of FERC’s authority over the electric sector, and EPA’s effort to assert authority through a statute that does not directly address these issues is inconsistent with Congress’ actions over nearly eighty years. 90 Pac. Gas & Elec., 461 U.S. at 205-06. 91 Energy Supply and Environmental Coordination Act of 1974, 15 U.S.C. § 792. 92 See Power Plant and Industrial Fuel Use Act of 1978, 42 U.S.C. §§ 8301, et seq. (repealed 1987). 93 Will v. Michigan Department of State Police, 491 U.S. 58 (1989). See also Altria Group, Inc. v. Good, 555 U.S. 70, 77 (2008) (providing a concise summation of the fund amental principle of statutory construction that was articulated in Will). 82 Cf. City of Arlington, Tex. v. FCC, 133 S. Ct. 1863, 1873 (2013) (finding objections regarding 94 “traditional state and local concern” inapposite where a federal statute “explicitly supplants” state authority). 95 CAA § 111(b)(1)(A), 42 U.S.C. § 7411(b)(1)(A). 96 Endangerment and Cause or Contribute Findings for Greenhouse Gases Under Section 202(a) of the Clean Air Act, 74 Fed. Reg. 66,496 (Dec. 15, 2009) [hereinafter “Endangerment Finding”]. 97 Proposed Rule at 34,841-44. 98 Section 202(a)(1), 42 U.S.C. § 7521(a)(1). 99 Section 111(b)(1)(A), 42 U.S.C. § 7411(b)(1)(A) (emphasis added). 100 See, e.g., Lowe v. SEC, 472 U.S. 181, 207 n.53 (1985); Sturges v. Crowninshield, 17 U.S. 122, 205 (1819) (“The fair, and we think, the necessary, construction of the sentence requires, that we should give these words their full and obvious meaning.”). 101 See EPA v. EME Homer City Generation, L.P., S. Ct. 1584 (2014); see also Michigan v. EPA, 213 F.3d 663 (2000); AFL-CIO v. Am. Petroleum Inst., 448 U.S. 607 (1980). 102 Am. Petroleum Inst., 448 U.S. at 655. 103 42 U.S.C. §§ 7408, 7409. 104 See 42 U.S.C. § 7545(c)(1) (endangerment finding for fuel additives); § 7547(a)(1) (endangerment finding for nonroad engines and nonroad vehicles required after an EPA study); § 7571(a) (endangerment finding for aircraft engines). See also 42 U.S.C. § 7603 (EPA’s emergency powers to file suits based on non-specified endangerment findings). 105 Endangerment Finding, 74 Fed. Reg. 66,496 (emphasis added). See also id. at 66,641, discussing “well-mixed” greenhouse gases with “common attributes” comprising the proper basis for the endangerment finding. 106 Endangerment Finding at 66517, 66519, 66524. 107 Proposed Rule at 34,842. 108 77 FR 22402 at n.32 and accompanying text. 109 79 FR 1455. 110 Section 111(d)(1)(A)(ii), 42 U.S.C. § 7411(d)(1)(A)(ii). 111 See Standards of Performance for New Stationary Sources; State Plans for the Control of Certain Pollutants From Existing Facilities, Final Rule, 40 Fed. Reg. 53,340 (Nov. 17, 1975) [hereinafter “1975 Implementing Regulations Preamble”] (“Section 111(d) requires control of 83 existing sources of such pollutants whenever standards of performance (for those pollutants) are established under section 111(b) for new sources of the same type.”). See also 40 C.F.R. § 60.22(a). 112 See Standards of Performance for Greenhouse Gas Emissions From New Stationary Sources: Electric Utility Generating Units, Proposed Rule, 79 Fed. Reg. 1430 (Jan. 8, 2014). 113 See Proposed Rule at 34,852; Legal Memorandum at 12 -13. It should be noted that the EPA applies to reconstructed as well as modified power plants. 114 Senate Committee on Public Works, S. Rep. No. 91 -1196, at 16-17 (1970). 115 42 U.S.C. § 7411(d)(1)(A)(ii) (emphasis added). 116 117 See 42 U.S.C. § 7411(a)(1). Essex Chem. Corp. v. Ruckelshaus, 486 F.2d 427, 433 (D.C. Cir. 1973). In addition, the Court in Sierra Club v. Costle, the court declined to remand EPA’s SO 2 and PM NSPS for EGUs in part because EPA had adequately examined the possibility that imposing baghouse technology on new and modified EGUs and concluded that doing so would not lead to costs that were so “excessive” that it would be “unreasonable” for EG Us to comply. 657 F.2d 298, 383 (D.C. Cir. 1981). 118 NERA, Potential Energy Impacts of the EPA Proposed Clean Power Plan (October 2014). 119 Id.at 21. 120 See, e.g., An Act Relating to Best System of Emissions Reduction For Existing Electric Generating Units, Kentucky House Bill 388 (signed April 2, 2014); Urge the United States Environmental Protection Agency, in Developing Guidelines for Regulating Carbon Dioxide Emissions from Existing Power Plants, to Respect the Primacy of Nebraska and Other States and to Rely on State Regulators to Develop Performance Standards, Nebraska Legislative Resolution 482 (adopted April 10, 2014). 121 NERA assumed that coal-fired power plants could achieve a maximum heat rate improvement of 1.5 percent, with more efficient plants a chieving a 0.75 percent heat rate improvement, and the most efficient plants achieving no improvement at all. NERA also assumed that heat rate improvements trigger New Source Review (NSR) for conventional air pollutants. NERA used a higher EE cost derived from academic literature. Alcott and Greenstone, Is there an Energy Efficiency Gap?, Journal of Economic Perspectives (2012). 122 NERA, Potential Energy Impacts of the EPA Proposed Clean power Plan. 21 (October 2014). 123 Id. 84 124 A more complete critique of EPA’s Building Block 1 assumptions can be found attached in Appendix 3 and 4. 125 Proposed Rule, 79 Fed. Reg. at 34,860. 126 Ed Cichanowicz and Michael Hein, Critique of EPA’s Statistical Evaluation Defining Feasible Heat Rate Improvements, 5 (December 1, 2014). 127 Id. 128 Id. 129 Id.at 6. 130 Id. 131 Id. 132 Id. 133 Id at 8. 134 Id. 135 Sargent & Lundy, Coal-Fired Power Plant Heat Rate Reductions (January 22, 2009). 136 Id. 137 Letter from Raj Gaikwad, Ph.D., Vice President for Advanced Fossil Technologies, Sargent & Lundy, LLC to Mr. Rae Cronmiller, Sr. Principal Environmental Counsel, National Rural Electric Cooperative Association (Oct. 15, 2014). 138 EPA, Technical Support Document (TSD) for Carbon Pollution Guidelines for Existing Power Plants: Emission Guidelines for Greenhouse Gas Emissions from Existing Stationary Sources: Electric Utility Generating Units GHG Abatement Measures , 2-32 (June 10, 2014). 139 J. Edward Cichanowcz, Critique of EPA’s Use of Reference Units to Select Heat Rate Reduction Targets (October 13, 2014). This critique addresses 14 of the 16 “reference units.” The remaining two units are being evaluated separately by the units’ owners and comments will be provided accordingly. 140 Id. at 1. 141 Id. at 2. 142 See Ed Cichanowicz and Michael Hein, Critique of EPA’s Statistical Evaluation Defining Feasible Heat Rate Improvements, 4 (December 1, 2014) (noting that heat rate goes up as capacity factor goes up). 143 Utility Air Regulatory Group v. EPA, __ U.S. __, (2014). 85 144 42 U.S.C § 7479(3), 145 EPA, New Source Review: Report to the President, 16 (June 2002). 146 A more complete critique of EPA’s Building Block 2 assumptions can be found attached in Appendix 5 and 6. 147 Energy Ventures Analysis, Redispatching Gas Combined Cycle Units, 10 (November 2014). 148 Id. 149 Id, at 11. 150 EPA, “Proposed Clean Power Plan_Option 1 Regional_ssr.XLS”. 151 EPA, “Proposed Clean Power Plan_Option 1 State_ssr.XLS”. 152 See generally NERC Initial Reliability Review at 9 -10. 153 See NERC Initial Reliability Review at 10. See, 154 e.g., NERC, Polar Vortex Review (Sept. 2014), available at http://www.nerc.com/pa/rrm/January%202014%20Polar%20Vortex%20Review/Polar_Vortex_ Re view_29_Sept_2014_Final.pdf; FERC & NERC, Arizona-Southern California Outages on September 8, 2011: Causes and Recommendations (Apr. 2012), available at https://www.wecc.biz/Administrative/FERC%20NERC%20Joint%20Report%20Arizona Southern%20California%20Outages%20on%20September%208,%202011.pdf . 155 See Energy Ventures Analysis, Redispatching Gas Combined Cycle Units, 10 (November 2014) (noting that historically, NGCC capacity factor varies between 12.6 percent and 63.4 percent, with most capacity factors ranging from 30 percent to 50 percent). 156 ACCCE, Recent Electricity Price Increases and Reliability Issu es Due to Coal Plant Retirements (Feb 6, 2014). 157 See Energy Ventures Analysis, Redispatching Gas Combined Cycle Units, 38 (November 2014) (noting that the pipeline infrastructure in the Western United States will limit the availability of increased NGCC generation). 158 James Marchetti, Review of EPA’s State Specific CO 2 Emissions Rate Goals: Building Blocks 2 and 3, 2 (November 2014) 159 Id. 160 Id. at 4-5. 161 Id. at 6 162 Id. at 13-16 86 163 Id. 18-19. 164 Id. 18-19. 165 EPA, Carbon Pollution Emission Guidelines for Existing Stationary Sources: Electric Utility Generating Units Notice of Data Availability , 79 Fed. Reg. 64,543 (October 30, 2014). 166 October 2014 NODA, 79 Fed. Reg. at 64,549 -50. 167 See, e.g., Sierra Club v. EPA, 499 F.3d 653, 655 (2007); Longleaf Energy Associates, LLC v. Friends of the Chattahoochee, Inc., 681 S.E.2d 203 (Ga. App. 2009); EPA Draft New Source Review Workshop Manual at B.13 (1990), available at http://www.epa.gov/ttn/nsr/gen/wkshpman.pdf (“Historically, EPA has not considered the BACT requirement as a means to redefine the design of the source when considering available control alternatives. For example, applicants proposing to construct a coal-fired electric generator, have not been required by EPA as part of a BACT analysis to consider building a natural gas-fired electric turbine although the turbine may be inherently less polluting per unit product (in this case electricity).”). 168 See 42 U.S.C. § 7411(a)(6) (defining “existing source” as “any stationary source other than a new source.”). 169 Proposed Rule, 79 Fed. Reg. at 34,857. 170 GHG Abatement TSD at 6-9. 171 40 C.F.R. § 60.22(b)(5). 172 October 2014 NODA, 79 Fed. Reg. at 64,550 -51. 173 A more complete critique of EPA’s Building Block 3 assumptions can be found attached in Appendix 6 and 7. 174 Energy Ventures Analysis, Review of EPA Building Block #3A – Renewable Energy, 7 (December 2014). 175 Id. 176 Id. 177 Id. at 8. 178 EPA, Technical Support Document (TSD) for Carbon Pollution Guidelines for Existing Power Plants: Emission Guidelines for Greenhouse Gas Emissions from Existing Stationary Sources: Electric Utility Generating Units GHG Abatement Measures , 4-11-15 (June 10, 2014). 179 Energy Ventures Analysis, Review of EPA Building Block #3A – Renewable Energy, 8-9 (December 2014). 87 180 181 Id. Department of Energy, Utility-Scale Land-Based 80-Meter Wind Maps, available at: http://apps2.eere.energy.gov/wind/windexchange/wind_maps.asp. 182 Energy Ventures Analysis, Review of EPA Building Block #3A – Renewable Energy, 10 (December 2014). 183 Id. 184 Id. 185 Alternative RE Approach TSD at 1-2. 186 Alternative RE Approach TSD at 2. 187 Alternative RE Approach TSD at 2. 188 For example, EPA arbitrarily assumes that the cost of RE technology will be reduced by up to $30/MWh. EPA cites no evidence to support this assumption, nor does it explain how this assumption affects its projection of the economically available RE developm ent potential. See Alternative RE Approach TSD. 189 From EPA’s Alternative RE Approach Technical Support Document: ID, IA, KS, NE, NM, ND, and SD. 190 It is important to note that beyond the methodological errors in EPA’s approach to setting the Building Block target, EPA erroneously concluded that the Pennsylvania study supports a conclusion that 1.5% annual increase, when it is in fact lower. Nexant, Review of EPA Clean Power Plan Building Block 4 – Energy Efficiency, 5 (November 25, 2014). A more complete critique of EPA’s Building Block 4 assumptions can be found attached in Appendix 8 and 9. 191 EPA, Technical Support Document (TSD) for Carbon Pol lution Guidelines for Existing Power Plants: Emission Guidelines for Greenhouse Gas Emissions from Existing Stationary Sources: Electric Utility Generating Units GHG Abatement Measures, 5-46-48 (June 10, 2014), Energy Ventures Analysis, Assessment of EPA Building Block 4 – Energy Efficiency (December 2014) 192 Energy Ventures Analysis, Assessment of EPA Building Block 4 – Energy Efficiency, 7 (December 2014). 193 Id. at 8. 194 Id. 195 Id. at 8-10. 196 Id. at 11-12. 88 197 Id. 198 Id. 199 Nexant, Review of EPA Clean Power Plan Building Block 4 – Energy Efficiency, 7 (November 25, 2014). 200 201 Id. at 5. Energy Ventures Analysis, Assessment of EPA Building Block 4 – Energy Efficiency, 7-8 (December 2014). 202 Id. 203 Nexant, Review of EPA Clean Power Plan Building Block 4 – Energy Efficiency, 7 (November 25, 2014). 7 204 Id. at 8. 205 Id. at 7. 206 North American Electric Reliability Corporation, Potential Reliability Impacts of EPA’s Proposed Clean Power Plan, 2 (November 2014). The full NERC report can be found attached as Appendix 10. 207 Id. 208 Id. 209 Southwest Power Pool, SPP’s Reliability Impact Assessment of the EPA’s Proposed Clean Power Plan (October 8, 2014). 210 Midcontinent Independent System Operator, Analysis of EPA’s Proposal to Reduce CO 2 Emissions from Existing Units (November 12, 2014). 211 Electric Reliability Council of Texas, ERCOT Analysis of the Impacts of the Clean Power Plan (November 17, 2014). 212 American Electric Power, Transmission Challenges with the Clean Power Plan (September 2014). 213 EPA, Regulatory Impact Analysis: Final Rulemaking for 2017-2025 Light-Duty Vehicle Greenhouse Gas Emission Standards and Corporate Average Fuel Economy Standards (August 2012). 214 ACCCE, Climate Effects of EPA’s Proposed Carbon Regulations , June 2014. 215 ACCCE, Re: Technical Update of the Social Cost of Carbon for Regulatory Impact Analysis Under Executive Order 12866 (February 26, 2014). 216 Id. 89 217 EPA, Regulatory Impact Analysis for the Proposed Carbon Pollution Guidelines for Existing Power Plants and Emissions Standards for Modified and Reconstructed Power Plants (June 2014); NERA, Potential Energy Impacts of the EPA Proposed Clean power Plan (October 2014). 90