Larry Tribe is stirring up trouble, and to do it, he’s relying in part on an argument I made (and Justice Kennedy seemingly adopted) in King v. Burwell. But Tribe’s King isn’t at all the same as my King. (And we don’t yet know what the Court’s King will look like.) Here’s the story.
In arguments before the D.C. Circuit on April 16, Tribe argued on behalf of several private and state intervenors that the EPA’s proposed Clean Power Plan exceeds the agency’s jurisdiction. The proposed rule would require states to enforce tough limits on their power plants’ carbon dioxide emissions, but if any state didn’t want the regulatory task, the Clean Air Act (CAA) would allow that state to turn the regulatory regime over to federal regulation within the state’s borders. In addition to arguing that EPA lacks jurisdiction to regulate power plants under the relevant statutory provision (CAA § 111(d)), Tribe argues that the proposed rule would violate the anti-commandeering doctrine by giving states a Hobson’s choice; either they must incur extreme regulatory costs themselves, or they must turn over their power systems to an undefined federal regulatory scheme that is, practically speaking, very unlikely to function well. No rational state, Tribe argues, would choose the unknown and likely dysfunctional federal scheme, especially in a realm as important and sensitive as electric power.
In testimony before the House Energy and Commerce Committee and in an online debate with his Harvard Law School colleagues Jody Freeman and Richard Lazarus, Tribe has analogized his anti-commandeering argument to the anti-coercion questions that Justice Kennedy asked in King (questions that seemed to arise from my amicus brief). Tribe’s point, as far as I can tell, is that the federal government is threatening the states with a federal fallback option that is likely to wreak havoc on the states’ energy systems.
As Tribe put it:
The very feasibility of a federal plan is open to question, because EPA has not announced a model and obviously lacks sufficient resources to operate throughout the nation all the utility regulatory programs, energy efficiency programs, or any of the other programs that would be necessary to implement a “federal” plan. And a poorly administered federal plan or one plagued with jurisdictional uncertainties and litigation could paralyze a State’s energy sector, damage its economy, and inflict serious hardship on its citizens. That’s why I see this dispute as strikingly similar to the Affordable Care Act case of King v. Burwell, argued in the Supreme Court on March 4. There, Justice Kennedy, among others, noted the “serious constitutional problem” that would result if a federal statute were interpreted as threatening the citizens of a State with significant injury unless the State agreed to follow federal policies. The same is true here.
In a Volokh Conspiracy post published shortly after the King arguments, Jon Adler cast the constitutional difficulty with the CAA in similar terms:
If thirty states refused to adopt SIPs, the CAA’s regulatory program would grind to a halt, as the EPA lacks the ability to assume that much responsibility for CAA implementation. And this would have exceedingly severe consequences for private firms in resistant states, as the CAA requires many firms to obtain operating permits – permits the EPA would lack the resources and staff to issue. So it’s simply false to say that Congress would never enact a statute that could be hamstrung by state resistance. It unquestionably has. What’s never happened before is this many states refusing to hum the government’s tune.
But as Brian Beutler recently argued in a piece at the New Republic, Tribe’s and Adler’s argument bears very little resemblance to the anti-coercion concerns that I raised in King. In the Clean Air Act, Congress has offered states the choice to regulate themselves or to have the federal government regulate for them, but the substantive legal rules are the same whether the state or federal government does the job. Tribe’s argument is not that the federal fallback is substantively or legally punitive; it is only that the federal government will perform so incompetently that the on-the-ground effect of a state’s noncooperation will be punitive. In King, by contrast, the point is not that the federal exchanges might operate less well than the state exchanges despite following identical rules; it’s that, under plaintiffs’ interpretation, the federal government would change the substantive legal rules in states that refused to set up their own exchanges. And the rules for noncooperative states would be intentionally, purposefully destructive.
In short, Tribe’s argument hinges on federal incompetence while mine hinges on federal maliciousness. And, constitutionally, that difference makes a huge difference.
Indeed, this distinction is central to all of American constitutional law, not just federalism. We don’t have a constitutional entitlement to a high-functioning government, as poor Joshua DeShaney discovered in 1989 and as the current Congress demonstrates daily. We do, however, have a constitutional entitlement to be free from governmental abuse, which is why we don’t simply shrug at the Ferguson report and why a North Charleston officer is standing trial for murder. In general, the government may do big things badly, but it may not do mean things wantonly.
In the federalism context, this distinction carries added importance because its collapse would hamstring the federal government far beyond the intentions of even the most ardent federalist and because its collapse would involve the judiciary in an inquiry that no true conservative wants to hand over to federal judges (making the argument unlikely to attract many conservative fans in the long run). Again, Tribe’s principle is that a threat of well-intentioned but probably incompetent federal regulation is unconstitutionally coercive of state governments. But this principle could not be cabined to cooperative federalism. Any time the federal government gets involved in a regulatory regime, it alters the landscape in a way that states might not like. Indeed, states frequently adjust federal regulatory regimes through exercises of their own regulatory machinery—a task that is often constitutionally possible even within limits of preemption. Under Tribe’s argument, those exercises of the states’ regulatory machinery could be considered “coerced” by the presence of federal regulation (or even just by the threat of federal regulation, while a bill was pending in Congress, for instance) if a court concluded that the federal regulation was sufficiently incompetent—or even just sufficiently likely to be incompetent—that a rational state would not want to leave the federal regulations unsupervised. That rule would be crazy on its face, but it would also involve the courts in scrutinizing the success of, rather than merely the justifications for, a federal law. That’s an inquiry that American courts have scrupulously avoided.
In short, a “threat” of well-intentioned federal regulation cannot be taken as a threat at all, even if litigants or states suspect that the federal government’s program will malfunction. The only way that a threat of federal regulation should be taken as a true threat is if it is clearly intended to be punitive or destructive, as the Obamacare threats at issue in King would be under the plaintiffs’ interpretation. And the Court cannot and will not impute to Congress (or the EPA) a subversive intent to regulate poorly within a facially well-intentioned regulatory program. As Justice Kennedy said at a recent congressional hearing, “We [the Court] have to assume that we have three fully functioning branches of government [ ] that are committed to proceed in good faith and with good will toward one another to resolve the problems of this Republic.”
All of this said, I think there are some serious federalism questions that one can ask about the Clean Power Plan. The regulatory burden that EPA proposes looks awfully big, and the inclusion of greenhouse gases under the CAA was arguably unforeseeable at the time of the statute’s passage, making this expansion of the CAA arguably analogous to the Medicaid expansion that the Court deemed unconstitutionally coercive in NFIB. Furthermore, the CAA authorizes EPA to punish states, through deductions in highway funding and stricter emissions offsets, if they refuse or fail to regulate. It would be possible—I might even argue advisable—for the Court to enjoin the imposition of those sanctions on states that choose federal over state regulation under the Clean Power Plan. But you don’t need the King argument to make that point; as far as that argument goes, it’s a straightforward application of NFIB. (And, notably, Tribe filed his opening brief in the CAA litigation, arguing anti-coercion, before I had written or filed my brief in King and long before Kennedy raised my anti-coercion concerns at oral argument.)
The only (farfetched, I think) King argument available against the Clean Power Plan is the following: EPA’s proposal would require states to regulate under § 111(d) even though EPA already regulates power plants under § 112. If there were evidence that EPA was actually regulating poorly and destructively under § 112 out of an actual intent to spur state participation, then I might argue for judicial involvement under the anti-coercion constraint, under logic similar to my King argument. And maybe a concern about this kind of deviousness could bolster Tribe’s argument that the CAA prohibits EPA from regulating sources under § 111(d) that it’s already regulating under § 112. But that’s not Tribe’s argument.