EPA and Global Carbon: Unnecessary Risk
This is the second post in a debate between RFF and Institute for Policy Integrity scholars over the best tools for EPA carbon regulation under the Clean Air Act. See the opening salvo from Jason Schwartz, and in particular Policy Integrity’s recent petition to the EPA. -ed
Jason Schwartz and Michael Livermore at Policy Integrity are first-rate Clean Air Act scholars and I’m honored to have the chance to debate with them. We agree on the most important points – above all that the EPA has ample authority under the existing statute to craft real, flexible, and effective carbon regulation. We do disagree some over the best way to get there. For some, this seems like legal arcana, but it matters. EPA’s choice of regulatory tool will do more than any other decision to shape the environmental ambition, cost-effectiveness, and legal vulnerability of its carbon regulation.
As Jason discussed, Policy Integrity recently petitioned the EPA to write comprehensive, flexible carbon regulation. On this, again, we agree. But Policy Integrity suggests EPA’s first choice of regulatory tool should be §115 of the Clean Air Act, which is aimed at international air pollution. This idea has been around for a while, at least since it was proposed by former EPA general counsel Roger Martella in a 2009 paper. Policy Integrity’s petition relegates the most discussed and mature approach – performance standards for stationary sources under §111 – to a third-best option. They argue – and I agree – that the two approaches are complements, not substitutes, but I think it would be a big mistake for EPA to ground its carbon regulation primarily in §115, rather than the much more robust §111. Doing so invites unnecessary legal risk.
The reason is simple: I don’t think Congress granted EPA broad regulatory authority in §115.
Instead, I think it only complements broad authority under other parts of the statute. §115 is very short and skeletal, at only about 300 words, almost all of that devoted to how it is triggered, not what actual authority it grants. Nothing in §115 limits how EPA (and the states) can regulate in terms of scope, stringency, or flexibility. Lack of boundaries is nice if you want EPA to have an open-ended mandate to write flexible regulations. But in the context of the statute, I don’t think it’s plausible that Congress intended §115 to be such a broad grant. As Justice Scalia puts it—and this isn’t a particularly “conservative” or unusual view—Congress does not “hide elephants in mouseholes”.
The problem isn’t that there’s an elephant (a broad delegation of authority) in a mousehole (a short, thin part of the statute) – it’s the hiding. Legislatures can certainly make broad delegations without much statutory detail. California’s A.B. 32 law is an excellent example of this—it granted state regulators sufficient authority to build a comprehensive cap-and-trade program without giving much specific direction (although it does give regulators much more guidance than §115 does). But Congress carefully created a variety of regulatory programs under the CAA for different sources and pollutants, each spelling out much more precisely than §115 does what EPA can and must do. It’s hard to imagine that Congress broke from this approach and granted almost unlimited discretion over both stringency and structure of regulation to address international pollution problems without being explicit—especially when other tools (like §111, or Title II for vehicles, etc.) are available for addressing domestic harms from the same pollution. In short, Congress could have given EPA carte blanche to regulate international pollution problems, but it’s not plausible to me that they intended to do so in §115, given the context of the statute.
Technical legal point: I think a court, using only traditional tools of statutory interpretation, could conclude the statute is not ambiguous here. In other words, I think EPA loses on Chevron step one if it argues for broad authority under §115.
If §115 doesn’t grant comprehensive authority, then why is it there at all? What was Congress trying to do? I understand it as a special tool for dealing with international harms that aren’t handled adequately by the core stationary-source regulatory programs (air quality standards under §108-110, toxic pollutants under §112, and performance standards under §111). Whatever those programs do based on domestic harms alone, if they’re inadequate to address harms outside the US (and you have reciprocity), you can crank up their stringency using §115. In this sense, §115 is the international analog to the tools available for dealing with interstate/downwind pollution (§110(a)(2)(d)(i)). That’s important, but it’s not the basis for an independent regulatory program. It should also inspire some caution to note that much of EPA’s recent trouble in the D.C. Circuit has come from its efforts to creatively use those analogous interstate provisions to build flexible regulation – moves that the court has repeatedly rejected, narrowly interpreting EPA’s powers to regulate based on interstate harms.
§115 is a good illustration of the foresight of the drafters of the Clean Air Act—it shows they were aware of international pollution problems, and aware that the major regulatory tools in the statute did not adequately address those problems. For this reason it’s a good counterargument to critics’ claims that the statute was never intended to address global pollution problems, and is incapable of doing so. But it doesn’t mean that the drafters intended §115 to be a separate, extraordinarily broad regulatory program. Instead, it’s much better understood as an accessory to the main regulatory tools.
In the GHG context, §115 remains very useful even under this narrower view. It allows EPA to consider harms from climate change suffered in foreign countries (at least those that give the U.S. reciprocity) when setting the stringency of GHG regulations including performance standards for power plants and fuel economy standards for vehicles. This is a big deal. As a colleague pointed out to me, social cost of carbon estimates are based on global harms, but EPA might only be able to consider domestic harms when regulating under §111 or Title II. §115 allows it to consider some foreign harms too, perhaps by enough to make the social cost of carbon a reasonable proxy on which to base benefit estimates. This means stricter GHG limits are legally justifiable.
Most of Policy Integrity’s petition is devoted to analysis of whether §115 can be used now for carbon – whether other countries are regulating it and giving the U.S. necessary reciprocity. Their analysis is spot-on here, and because I think §115 plays this important role, it’s very welcome. In short, §115 can and does apply to GHGs. But that doesn’t lead to the conclusion that the EPA can or should use §115 as the foundation of its carbon regulation. I don’t think the petition is convincing here.
§111 performance standards, in contrast, are well on their way and are on much stronger legal footing as the primary regulatory vehicle. In fact, EPA has already promised §111 standards for the most important sectors in settlement agreements, and will be legally required to take the most important step – standards for existing power plants – as soon as proposed standards for new sources are finalized. Along with others, including Jason and Michael at Policy Integrity, I’ve been studying how EPA can and should write these standards for some time. My reading is that EPA can write broad, flexible regulations here that belie the reputation of the Clean Air Act as rigid and costly. EPA should stay the course here, and consider using §115 to augment these standards in the future.
To play devil’s advocate, many of the same criticisms I direct toward §115 could apply to existing-source standards under §111(d) – that part of the statute is short and infrequently used too. It’s true that there is real legal risk. But I think that risk is much less than if a similar program were based on §115, though. There’s more to §111 than §115, in terms of both substance and procedure.
Flexible, multisector GHG regulation is a big castle to build. §115 is a nice sandbox. You can make a beautiful castle with it. But it’s still made of sand.