Join us.

We’re working to create a just society and preserve a healthy environment for future generations. Donate today to help.


The Clean Power Plan: Unpacking the Generation Shifting Issue

Climate Justice

The U.S. Environmental Protection Agency’s (EPA’s) Clean Power Plan (CPP) relies, in part, on a pollution reduction strategy – generation shifting – that is at issue in the ongoing lawsuit over the rule. Generation shifting involves increasing use of relatively clean natural gas and renewable energy and reducing use of relatively dirty and expensive coal-fired power plants. Although the technique has lowered power plant emissions significantly in recent years, opponents of the CPP have argued in legal briefs that section 111 of the Clean Air Act precludes relying on generation shifting to reduce carbon dioxide emissions. They claim the technique somehow does not lead to reductions at a pollution source, but their argument doesn’t hold up under scrutiny. Generation shifting does reduce emissions at each pollution source that takes advantage of the technique and therefore passes muster. 

Some explanation of section 111 and the CPP will set the stage for analyzing this issue. Subsection 111(d) requires states to adopt “standards of performance” for existing sources of carbon dioxide (CO2). Subsection 111(a) defines a standard of performance as a “standard for emissions of air pollutants which reflects the degree of emissions limitation achievable through the application of the best system of emission reduction which . . .” EPA “determines has been adequately demonstrated.” (emphasis added). 

EPA’s CPP offers guidance to states on how to develop the performance standards that section 111(d) requires. The CPP does not itself establish standards of performance for individual sources. Instead, the CPP guides states by using achievable emission rates for individual facilities as a guide to establishing limits on CO2 emissions for the electric utility sector as a whole in each state, leaving to the states the responsibility of promulgating individual performance standards. The EPA’s goals for state utility sectors rely in part on the feasibility of generation shifting. That is, EPA has concluded that the best system of emission reduction for power plants includes generation shifting. 

Is Generation Shifting a System of Emission Reduction under Section 111(a)? 

The term “system of emission reduction” literally embraces any technique that reduces emissions of air pollution. Generation shifting reduces emissions at regulated power plants by reducing their hours of operation and thereby reducing their emissions. So, the text of section 111 quite clearly allows generation shifting. 

The electric utility owners and states challenging the CPP at the U.S. Court of Appeals for the D.C. Circuit (hereinafter polluters) seek to read into the statute a prohibition on relying on reduced hours of operation or shutdowns of dirty facilities as systems of emission reduction. Emissions can come down because a facility installs an end-of-pipe control, uses cleaner fuel than in the past, or reduces its activity rate (e. g. hours of operation). No language in the statute privileges the utilities’ preferred end-of-pipe control technique over fuel switching or reduced activity rates. Indeed, the Pollution Prevention Act (passed in the same year as the 1990 Clean Air Act Amendments) expresses a preference for “source reduction” defined as “any practice which . . . reduces the amount of any. . . pollutant . . . released into the environment” over end-of-pipe control.1 

A policy logic does underlie the polluters’ preference for end-of-pipe control. It seems intrusive to demand the sort of fundamental technological changes called for in the Pollution Prevention Act. But in the states that have demanded greenhouse gas reductions, utilities have voluntarily engaged in generation shifting rather than more costly end-of-pipe control. 

The better economic rationale for disfavoring limiting production at a facility as a pollution control technique involves concern that doing so might deprive some consumers of the products the facility makes. This rationale, however, does not apply to generation shifting. Generation shifting changes how electricity is produced, not the total quantity supplied. 

May EPA Provide Guidance for the Utility Sector as a Whole? 

EPA’s guidance to states establishes goals for the utility sector as a whole based on emission rates achievable at individual facilities, rather than precise requirements for each power plant. The Clean Air Act contains no restrictions on EPA guidance for state submissions under section 111(d). Section 111(d) only explicitly requires EPA to “establish a procedure” for state submission of a state plan to establish the performance standards required by the section.2 

EPA has for decades interpreted the statute as authorizing guidance on the “best system of emission reduction.”3 In the absence of any specific statutory restriction on the sort of guidance the agency can offer, it seems appropriate to provide a goal based on average limitations achievable through a system of emission reduction instead of having EPA itself figure out specific limits for each power plant in the country. After all, the states have the responsibility to establish performance standards under section 111(d)(1), subject to EPA review and approval. 

Does Generation Shifting Preclude Submission of a State Plan Establishing Standards of Performance for Existing Sources? 

The polluters argue that incorporating generation shifting precludes state establishment of emission limits for pollution sources that section 111(d) requires. That argument amounts to an arbitrary and capricious claim disagreeing with EPA’s conclusion that states can implement the rule with facility-by-facility limits. EPA’s rule contains a provision authorizing submission of state rules for individual facilities.4 Hence, it appears that the agency believes that it is possible to implement source-specific standards under the rule it has designed. 

This belief makes sense even when states rely on generation shifting as a technique for achieving the rule’s goals. After all, generation shifting reduces the emissions at polluting facilities by reducing their hours of operation because of greater availability of cleaner generation. Thus, translating generation shifting into individual limits requires forecasting the effect of generation shifting on the emissions at dirtier plants or even committing to reduced hours of operation at dirtier plants based on pollution control considerations and new generating capacity. Of course, if states choose to shut down dirty old facilities because of growth of newer, cleaner generation, the shuttered facility will predictably meet a zero-level emission limit. 

The rule’s challengers have not advanced any reason why states cannot technically establish emission limits for individual facilities based on generation shifting. Instead, they have argued that generation shifting conflicts with the requirement that states establish standards of performance “for” individual sources. But section 111(a)(1) defines a standard of performance as “a standard of emissions of air pollutants.” Hence, as long as states write limits on power plant emissions, they have developed standards of performance – i.e. emission limits – “for any existing source,” as the Clean Air Act requires.5 

The polluters instead interpret the requirement that a performance standard apply to existing sources as precluding paying for off-site activity in conjunction with meeting the emission limit. When an electric utility operator complies with a performance standard by reducing hours of operation at his regulated power plants, he will likely make it possible to reduce dirty generation without harming supply by increasing generation at cleaner facilities. All emission reduction standards require polluters to pay for some service off-site, such as manufacture and delivery of a pollution control device or production and shipment of cleaner fuel. But a performance standard by itself only establishes a numerical emission limit, which does not limit the techniques that owners of facilities may use to comply. 

In short, a performance standard applies to and is for a source when it limits the source’s emissions. The requirement to apply performance standards to dirty power plants does not limit the techniques electric utilities may use to comply or that EPA may consider in setting requirements. Cf. 42 U.S.C. §7411(h). 

Furthermore, EPA has previously interpreted requirements that apply to “any source” to allow the use of generation shifting. Section 110(a)(D)(i) requires states to prohibit “any source or other type of emissions activity within the state” from emitting pollution interfering with other states’ achievement of air quality standards. EPA’s Cross-State Air Pollution Rule complied with this provision by creating an emissions trading program based in part on generation shifting.6 Similarly, EPA implemented a requirement that “each major stationary source” install the “best available retrofit technology” (BART),7 by authorizing state trading programs, in part to take advantage of the cost effectiveness of generation shifting. 64 Fed. Reg. 35,714, 35742 (July 1, 1999) (justifying allowing trading as a means of implementing BART because it allows source owners to take changes in fuels and processes into account “across . . . sources”).   

Do the Compliance Flexibilities Provided by EPA Conflict with the Requirement to Apply Performance Standards to Existing Sources? 

The challengers’ argument against allowing the possibility of generation shifting to influence EPA’s goals for state performance standards seems off-target. Some provisions of the rule raise legitimate legal issues. But the polluters do not challenge those provisions directly, probably because doing so does not serve their interests.   

EPA has authorized emissions trading and some procedures for state demonstration of compliance that raise questions about their consistency with the requirement to apply performance standards to “any existing source.” The petitioners do not challenge these compliance flexibilities. Had they challenged them, the court would have to decide, for example, whether a state emissions trading program conflicts with the requirement that states write standards for “any existing source.” A small literature exists on this subject, with most commentators finding trading consistent with section 111(d). 

These flexibilities ease reliance on generation shifting, but the CPP does not require their use, suggesting that states can implement generation shifting without them. In any case, even a conclusion that some or all of the compliance flexibilities are illegal does not demonstrate that EPA may not base its guidance on generation shifting and that states cannot rely on this technology in figuring out achievable limits for their power plants. 


Section 111 of the Clean Air Act authorizes reliance on generation shifting as a “system of emission reduction.” The polluters have transformed plausible arguments about tensions between compliance flexibilities and the requirement that states write standards “for sources” into invalid arguments creating limits stated nowhere in the statute regarding the types of technological systems EPA may rely upon in guiding state performance standards under section 111(d) of the law.

1 42 U.S.C. §§ 13101, 13102.

2 42 U.S.C. §7411(d)(1).

3 See 42 U.S.C. §7411(a)(1).

4 40 C.F.R. §60.5740(a)(2).

5 42 U.S.C. §7411(d)(1).

6 76 Fed. Reg. 48,208, 48,252-3 (Aug. 8, 2001).

7 42 U.S.C. §7491(b)(2)(A).

Climate Justice

Subscribe to CPRBlog Digests

Subscribe to CPRBlog Digests to get more posts like this one delivered to your inbox.