What EPA should do with its delayed performance standards for GHGs
On September 15, EPA announced that it would not meet its September deadline for proposing performance standards for greenhouse gas (GHG) pollution from power plants. (That is the second delay; this proposal was originally scheduled for July 2011.) Some are asking if this delay is a big deal, and several environmental leaders sent President Obama a letter requesting the prompt release of the new standards.
These would be New Source Performance Standards under Section 111 of the Clean Air Act, which generally require efficiency upgrades for new or modified industrial sources of emission. Basically, EPA determines the feasible level of emissions reduction for a given pollutant from a given source (such as power plants). EPA sets a “performance standard” that reflects that level of emission reduction and requires new or modified sources (again, power plants in this case) to meet that standard. Then, through Section 111(d), each state defines a performance standard for all existing sources in the state that, again, reflects that level of emission reduction.
Given that carbon dioxide is a well-mixed global pollutant, EPA may have a lot of flexibility in defining the performance standard. Two interesting possibilities exist: (1) EPA could implement some sort of national trading program, whereby power plants could trade allowances for GHG emissions in order to meet the required level of emissions reduction. (2) EPA could allow states—such as California—to use a trading program to meet their Section 111(d) requirements. The latter possibility is admittedly more feasible (and likely) than the former.
In any event, EPA should take advantage of its delays by coming up with an innovative method to regulate GHG emissions under Section 111. At least then EPA could demonstrate that it was not just delaying its proposed regulations for the sake of delay. Below, I go “into the weeds” a bit and highlight the potential for each option.
EPA previously tried to establish a trading program for mercury emissions under Section 111 in its Clean Air Mercury Rule (CAMR). EPA ran into legal difficulty with CAMR, however, because mercury is a hazardous air pollutant traditionally regulated under Section 112. In New Jersey v. EPA (2008), the D.C. Circuit threw out CAMR because EPA was trying to regulate mercury under the wrong section of the Clean Air Act. The D.C. Circuit thus never reached the legal argument discussed in the briefs to the court on whether EPA had authority to establish an emissions trading program under Section 111.
In theory, EPA could try again, this time for GHG emissions, which have not been classified as hazardous air pollutants. Using Section 111, EPA could define a standard of performance for GHG emissions such that power plants could meet the standard if they had sufficient allowances to cover their emissions. This definition would, in effect, establish a trading program. (The devil, as always, would be in the details.) The allowance standard would reflect the level of emission reduction required by Section 111 if EPA limited the total number of allowances available, resulting in lowered total emissions (establishing a “cap” on emissions).
A trading program that accomplished emissions reductions at power plants equal or greater to the required level of emission reduction as determined by EPA would arguably meet legal scrutiny. This “required level” is basically what EPA determines is feasible by conducting a technology review for the given industry. A trading program could do better than this because, frankly, Section 111 requirements can account for cost and are thus not generally set to the most stringent reduction achievable by the regulated sources.
Of course, EPA may not be ready to reintroduce a CAMR-style trading system for GHG emissions. Litigation, from both industry and environmental groups, would be a given. And EPA would be stuck defending a relatively new, untested program.
Of greater potential, then, may be the role of states like California. Under Section 111(d), each state defines a performance standard for existing sources based, in large part, on the level of emission reduction determined by EPA’s technology review. California is already set to regulate GHG emissions from power plants using its cap-and-trade program. (Similarly, RGGI also regulates GHG emissions from power plants.) Thus, it is in industry’s best interest to avoid being regulated twice: once by California cap-and-trade and once by Section 111(d) of the Clean Air Act.
EPA could therefore structure its regulation so as to encourage states, like California, to use participation in an existing trading program as a method to meet Section 111(d) requirements. To do this, California would need to demonstrate that the level of emission reduction from power plants under its cap-and-trade program would meet or exceed that EPA-determined level of emission reduction. California would then define the standard of performance to include retirement of emission allowances (and possibly increase the stringency of its program to meet EPA requirements). Ideally, a power plant that complies with California’s trading program would simultaneously comply with Section 111(d).
The legal catch is this: would any verifiable emission allowance “count” for section 111, if owned and retired by a regulated power plant? In other words, must all the emission reduction come from (a) the specific power plant; (b) the regulated power plant industry; or (c) any verifiable emissions source? The more narrow the scope of the required emission reduction, the more stringent the trading program must be to remain in compliance with Section 111. (Under (b), for example, only allowances tied to emission reduction from other power plants would “count” for Section 111 compliance.) The statutory language is less than clear on this point, and I would argue that either (b) or (c) is the correct answer in the case of GHG emissions.
Shameless self-plug: If you are dying to read more on trading under Section 111 or using California’s cap-and-trade program to meet Section 111(d) requirements, I have a paper on SSRN that does just that. http://ssrn.com/abstract=1879513.