This is the third in a series of posts offering some initial insights and observations, and posing several open legal questions related to EPA’s proposed 111(d) rule. (See the first and second posts.)
Over the course of this series, I welcome our knowledgeable and insightful LegalPlanet audience to join the dialogue in the comments. What strikes you about the proposed rule? What legal questions puzzle you? What are your thoughts on the below issues?
My first post made some big-picture observations about EPA’s proposed 111(d) rule, and my second post dug into EPA’s determination of the best system of emission reduction (BSER) and calculation of state goals. Now, I turn to some of the ways the rule may interact with existing greenhouse gas cap-and-trade programs and facilitate new regional collaborations.
EPA interprets the Clean Air Act term “standard of performance” to allow averaging and trading. Before diving into the nitty-gritty specifics of how the proposed rule might interact with state cap-and-trade programs, it is important to emphasize that EPA interprets §111(d) to allow state plans to group emissions from multiple affected electric generating units (EGUs) under a unified statewide target. As I have mentioned in my prior posts about the proposed rule, this interpretation is critical, and likely will be the focus of legal challenges to come. If each and every regulated EGU instead had to reduce its own emission rate or mass by a defined amount, there would be far fewer opportunities for cost-effective emission reduction–and existing cap-and-trade programs would be less effective in helping states comply with the rule. EPA recognizes that the most efficient and effective emission reduction program allows all units in the sector to average or trade their emissions.
EPA’s reasoning justifying this interpretation bears repeating here:
“The EPA proposes that the definition of ‘standard of performance’ is broad enough to incorporate emissions averaging and trading provisions, including both emission rate programs, in which sources may average or trade those rates, and mass emission limit programs, in which sources may buy and sell mass emission allowances . . . . The term ‘standard’ in the phrase ‘standard for emissions of air pollutants’ is not defined in the CAA. As the Supreme Court noted in a CAA case, a ‘standard’ is simply ‘that which ‘is established by authority, custom, or general consent, as a model or example; criterion; test.’’ A tradable emission rate or a tradable mass limit is a ‘standard for emissions of air pollutants’ because it establishes an emissions limit for a source’s air pollutants, and as a result, qualifies as a ‘criterion’ or ‘test’ for those air pollutants” (pp. 523-24). Furthermore, “an averaging or trading requirement qualifies as a ‘continuous emission reduction’ because, . . . in the case of a tradable mass limit, the source is always under the obligation that its emissions be covered by allowances” (p. 524).
EPA also refers to the two historical examples of §111(d) rules that did not follow a traditional emission rate limit. One of these examples, the Clean Air Mercury Rule, would have reduced emissions of mercury from power plants through a cap-and-trade program, but the D.C. Circuit overturned the rule on other grounds in 2008 (p. 524). EPA’s reasoning in the proposed CO2 rule harkens back to similar arguments the agency made in the Clean Air Mercury Rule litigation and foreshadows arguments that are likely to play a key role in future litigation.
You can expect opponents of the proposed rule to argue that the statutory term “standards of performance for any existing source,” in the broader context of the Clean Air Act, implies that a state plan must require each and every affected EGU to reduce its emissions mass–or even its emission intensity. Opponents may rely on ASARCO Inc. v. EPA, 578 F.2d 319 (1978), as they did in the Clean Air Mercury Rule litigation. According to this argument, averaging or trading programs do not guarantee emissions reductions from each and every participating unit, and therefore do not satisfy the statutory requirement.
The big-picture take-away: EPA’s interpretation that section 111(d) allows emission averaging or trading among affected EGUs provides the foundation for existing state-level cap-and-trade programs to play a role in compliance and facilitates the development of new cap-and-trade programs. This interpretation is important, and likely will be litigated. (See my Part I post for discussion of how the proposed rule would further allow states to benefit from policies like renewable portfolio standards and energy efficiency programs that reduce the overall emission intensity of the electricity sector.)
EPA chose not to amend its subpart B regulations. As described above, EPA interprets the Clean Air Act to allow affected EGUs to average or trade their emissions under a statewide budget. But do EPA’s regulations allow averaging or trading?
EPA’s implementing regulations at 40 C.F.R. part 60, subpart B interpret the Clean Air Act statutory requirements for state plans. Notably, the regulations interpret the statutory term “standard of performance” in the state plan context as “a legally enforceable regulation setting forth an allowable rate of emissions into the atmosphere, establishing an allowance system, or prescribing equipment specification for control of air pollution emissions.” Furthermore, the regulation must apply to all affected facilities in the state (40 C.F.R. § 60.21(f)). Multiple components of this definition could present challenges when applied to EPA’s proposal. Additionally, there is some uncertainty about whether the regulations’ reference to “an allowance system” is still a valid following the D.C. Circuit’s vacature of certain regulations in the Clean Air Mercury Rule litigation. Is a statewide emission budget a “legally enforceable regulation” (see my Part I post for more discussion)? Does a statewide budget set forth an “allowable rate of emissions?” Is the reference to “an allowance system” still valid? Does the requirement that standards of performance apply to all affected facilities mean that each and every EGU must reduce its emissions?
Prior to the release of this proposal, commentators speculated about whether EPA would open its subpart B regulations for amendment as part of the proposed rulemaking. The proposal indicates that EPA is not considering amendments to its implementing regulations (see pp. 29, 77). This decision may have important implications for future legal challenges to the guideline and state plan approvals. My initial reaction is, although EPA receives great deference in interpreting its own regulations, why not strengthen its legal position by amending the subpart B regulations? Does the risk of the legal challenges to subpart B regulatory amendments outweigh the impact of possible conflicts between the subpart B regulations and the proposed rule? Would regulatory changes negatively impact EPA’s ability to control other pollutants?
How should EPA categorize sources to facilitate emission trading? Having established that EPA interprets Clean Air Act section 111 generally to allow trading/averaging, it is now important to understand whether EPA needs to craft the rule in a particular way to facilitate trading/averaging. In particular, need EPA categorize all affected EGUs within the same source category?
EPA solicits comments on combining the two existing EGU source categories—fossil fuel-fired steam generating boilers and fossil fuel-fired combustion turbines—into a single category for the purposes of facilitating emission trading among affected EGUs (pp. 133-34). In the NSPS that EPA proposed in January, the agency suggested two alternatives in the context of new sources: keeping the categories distinct or creating a new, combined “TTTT” category. EPA solicited comment in the proposed NSPS on whether combining the categories for new sources is necessary in order to combine the categories for existing sources. Is a combined NSPS category necessary for a combined existing source category?
Now, in the context of existing sources, “EPA solicits comment on whether combining the two categories would offer additional flexibility,” e.g., by facilitating redispatch or emissions trading (p. 134-35). Additionally, “EPA solicits comment on whether combining the categories is, as a legal matter, a prerequisite for . . . facilitating averaging or trading systems that include sources in both categories, which states may wish to adopt” (p. 330). Is a combined existing source category essential to accommodate trading?
The proposed rule facilitates regional collaboration. The proposal contemplates that states will join together to submit multi-state plans to achieve multi-state goals, and encourages regional compliance. For states submitting a multi-state plan, EPA declares that individual state goals would be replaced with a multi-state goal (p. 116; see also p. 438). Any multi-state goal calculation, however, must account for the fact that the “mix of generation among affected EGUs in different states could differ significantly during the plan performance periods from that during the 2012 base year” (p. 439). For this reason, and to facilitate regional cap-and-trade programs, should EPA issue regional goals as well as state-specific goals? Or should EPA allow multi-state plans to propose multi-state goals grounded in the state-specific goals determined by EPA? Does the complicated modeling inherent in developing multi-state goals and plans present a major obstacle to state agencies interested in regional partnerships?
How far should the proposed rule go in promoting cap-and-trade? EPA notes that some stakeholders advocate for a regional trading program to constitute BSER. EPA specifically solicits comments on whether a regional trading program could/should constitute BSER (p. 331). Should/could a regional cap-and-trade program constitute BSER? (See my Part II post for more discussion of BSER.)
Additionally, EPA acknowledges requests from some stakeholders for a model rule outlining an interstate cap-and-trade program, but EPA has not proposed such a model in this document (p. 94). Would it be helpful to states for EPA to propose a model regional cap-and-trade program? Should EPA then issue mass-based targets for each state, or for multi-state regions?
The out-of-sector offsets included in most existing cap-and-trade programs would not count toward §111(d) compliance. EPA recognizes that emission trading programs such as California’s multi-sector cap-and-trade program and RGGI incorporate multiple flexible compliance mechanisms, including “multi-year compliance periods; the ability to bank allowances issued in a previous compliance period for use in a subsequent compliance period; the use of out-of-sector project-based emission offsets; and cost-containment allowance reserves” (pp. 498-99; see also p. 430). Consequently, the actual emissions from EGUs participating in a cap-and-trade program may exceed the cap. Additionally, California’s economy-wide cap-and-trade program extends beyond the electricity sector.
All of these mechanisms make cap-and-trade programs more cost-effective and flexible, but present a challenge in the context of §111(d), which focuses exclusively on reducing emissions from regulated sources in the electricity sector. EPA’s BSER is focused specifically on reducing emissions from coal- and natural gas-fired power plants. Therefore, EPA proposes that while state plans can include trading programs that incorporate flexible compliance mechanisms (e.g,. California’s A.B. 32 program), states must demonstrate that affected EGUs alone are achieving the required emission reduction without offsets (see p. 429). In other words, even if California’s EGUs use project-based offsets instead of electricity-sector emission reductions to meet their cap under A.B. 32, the total emissions intensity of California’s electricity sector still has to meet the federal target.
Thus, while states can use existing state cap-and-trade programs to meet the federal requirement, they cannot simply refer to an existing cap-and-trade program cap as ipso facto evidence that the state’s electricity sector is achieving the required reductions. I imagine California will engage in some sophisticated modeling to demonstrate that, under A.B. 32, California’s electricity sector is meeting the federal emission intensity target even without offsets and not counting out-of-sector reductions. See the State Plan Considerations TSD for further discussion of offsets (pp. 37, 40).
Stay tuned for more insights, observations, and legal questions related to EPA’s proposed section 111(d) rule.