Regulating Existing Power Plants Under Clean Air Act 111(d) (Part I): The CAA's Language & Structure

In late September, the EPA proposed regulation of new power plants’ greenhouse gas emissions (GHGs) under the Clean Air Act’s “New Source Performance Standards” (NSPS) provisions. Now an often little noticed follow-on provision—Section 111(d)--- is suddenly in the spotlight. Section 111(d) requires regulation of existing sources that are in categories of polluters subject to NSPS regulation. President Obama, EPA, industry, environmental groups, and states have all entered the fray about what Section 111(d) requires and allows. 

This issue presents several important choices and issues. First, regulation of existing US fossil fuel burning power plants—the source of over 30% of US carbon dioxide emissions-- is unavoidably central to US efforts to reduce GHG emissions. Second, in the face of the federal government’s long climate change inaction, many states and regions have already created programs and laws designed to reduce GHG pollution. Some have embraced variants on cap-and-trade schemes, or enacted renewable portfolio standards, or developed demand reduction strategies focused on other sectors, technologies, and consumer use. How would new Section 111(d) mandates mesh with diverse state efforts? Over the long term, Section 111(d) regulation could demonstrate to congress, states, industry and perhaps the world what is possible by providing a major testing ground and microcosm for more comprehensive climate regulation, whether at the national or international level. 

A major dividing line in early comments is whether states can adopt diverse strategies that consider or allow strategies such as pollution trading, or whether such regulation needs to focus “inside the fence line,” meaning regulatory limits no more stringent than could be derived through a more individualized unit or plant-based analysis. Much of the early commentary may just reflect taking of sides, but if taken seriously seems to show misunderstandings of the actual language of Section 111(d) and several crucially important cross references.

Section 111(d)’s key language states (italicization mine) that EPA “shall” issue regulations that “shall establish a procedure similar to that provided by section 110 under which each state shall submit to the Administrator a plan which (A) establishes standards of performance for any existing source for any air pollutant” subject to NSPS regulation. States must be permitted to consider “the remaining useful life of the existing source” in devising such a plan. What this exactly means requires a resort to the definition of a “standard of performance,” which is stated to be “a standard for emissions of air pollutants which reflects the degree of emission limitation achievable through the application of the best system of emission reduction which (taking into account the cost of achieving such reduction and any non-air quality health and environmental impact and energy requirements) has been adequately demonstrated.” Two other provisions have been generally overlooked, but are important to what EPA and the states have latitude to embrace. Section 110(a)(2)(a) in 1990 was amended to explicitly authorize use of tradable instruments and economic incentives in State Implementation Plans (SIPs). Second, the CAA explicitly provides in Section 116 that Section 111's requirements are a regulatory floor; states can mandate pollution reductions more stringent than what EPA might require.

Section 111(d)'s reference to Section 110 and call for a “procedure” and a state “plan” is of critical importance, as is the requirement that limitations “reflect ” the “best system” of reduction “adequately demonstrated.” Section 110 is the provision that empowers states to act as architects of plans--SIPS--to ensure that state regions meet national “ambient air quality standards.” 111(d) is not focused on those same pollutants, but is explicitly calling for a planning process and federal-state relationship as long established under Section 110.   And Section 110 in effect is one of the earliest variants of a “cap-and-trade” system. It requires each designated state region to look at the required level of air quality (the cap) and then devise plans to get there or maintain better than necessary air quality (with limited exceptions). As the US Supreme Court has confirmed, states are in the Section 110 SIP-planning driver’s seat. They can make their own choices and allocate burdens, with EPA’s role limited to one of review to ensure a state’s plan is adequate, real, enforceable, and is implemented and achieves goals as promised. (If a state flunks at one or more required stages, the federal government eventually can step in, as it can under Section 111(d).)

So what does Section 111(d)’s cross-reference to Section 110 mean when “standards of performance” are targeted at particular polluter categories, in this case, existing power plants?   It means at least an embrace of state primacy in choosing how to regulate particular individual power units or plants, but also encompasses a state planning process. And since Section 110’s SIP-planning is all about pollution caps and initial state latitude to devise strategies, that too seems a part of this cross-reference's meaning.   Furthermore, because Section 110 has since 1990 explicitly allowed use of marketable permits and economic incentives in devising SIPs, EPA and the states have a strong statutory basis for utilization of similar sorts of strategies in Section 111(d) plans. And “plans” and “procedures” need only “reflect” what the “best system" can achieve; it does not use ‘best technology” as the benchmark, as used in other provisions of the Clean Air Act and other environmental laws. So, putting aside for now whether the unit of analysis is an electricity generating unit, a plant, or perhaps even a state or region's carbon pollution, while a state will likely have to derive a sort of aggregate pollution budget based on what its regulated existing power generating sources can achieve under “systems” that are “adequately demonstrated,” nothing in this language mandates an “inside the fence” focus in how states could derive what that budget should be or in creating a "plan" to achieve required pollution reductions. Like Section 110, Section 111 is about results and broader flexible planning. 

Some comments that advocate this more unit-based focus (so called "inside the fence" regulation) seem to be claiming that 111(d) plans cannot impose more stringent reductions on power plants than some baseline level of pollution that EPA or the states would derive.   Due to Section 116 language allowing more stringent state regulation, this claim is undoubtedly in error. If a state wants to look at its coal burning plants and mandate pollution reductions that take into account trading cost efficiencies, or simply crack down on coal burners when drafting a plan required by Section 111(d), EPA can only reject such a plan if it is too lax. In EPA’s eventually released Section 111(d) regulations and linked guidance documents, it cannot prohibit states from doing more. 

All laws provide room for argument about their meaning. But under the famous Chevron case, EPA is the delegated and expert agency charged with Section 111(d) interpretation, with substantial judicial deference due.  And Sections 111 and 110 further grant planning discretion to states. The interpretive sweet spot here—meaning the most comfortable, natural read of these intertwined provisions---is that states are the initial planners of means to reduce GHG pollution from existing power plants, and that the degree of emission reduction such plans will achieve must be at least as stringent as “reflected” through assessment of the ”best system” of reduction demonstrated. And states can require greater stringency. EPA will have to demand viable, real, measurable and enforceable reductions, but beyond that, EPA regulation should provide states with the ability to adopt strategies that best meet their particular mix of sources, current GHG initiatives, and aspirations. 

Forthcoming blog analysis of Section 111(d) regulation will analyze in more depth the implications of the Chevron decision for this choice, different ways EPA and the states could derive both the amount of required pollution reductions and means to achieve those reductions (including more about the implications of Section 111(d) as generating only a regulatory floor), and, as a policy matter, the implications of strategies that might range from a unit-specific to state or regional focus.