Symposium: Economics beats formalism

Einer Elhauge is the Petrie Professor of Law at Harvard Law School

My major takeaways from the Supreme Court’s opinion in King v. Burwell are these:

First, this opinion confirms an interesting divide among Supreme Court conservatives between the economic conservatives and the formalist conservatives. Chief Justice John Roberts’s opinion begins with a delightfully lucid explanation of the economics of adverse selection problems that the dissent’s interpretation would create. (Is there anyone who can express economics in plain English better than Roberts?) He ends with the ringing statement: “Congress passed the Affordable Care Act to improve health insurance markets, not to destroy them. If at all possible, we must interpret the Act in a way that is consistent with the former, and avoids the latter.” In the middle, there is lots of disagreement about various textualist details (more on that later for the intrepid reader), but this general difference in outlook was key.

Not surprisingly for someone who was a prominent antitrust advocate in cases like Microsoft, Roberts has a sophisticated understanding of economics and his conservativism is more of the law-and-economics school variety. At least when it comes to economic regulation, he tries to read statutes to make economic sense rather than economic nonsense. Justices Antonin Scalia, Clarence Thomas, and Samuel Alito are pure formalists, happy to lead us all into economic disaster if they think that is the linguistically most accurate statutory reading. We saw the same divide in the past Obamacare case, in which Roberts recognized the lack of economic difference between a tax and a penalty and interpreted coercive threats in a way consistent with contract law and economics, as I have shown elsewhere. We also saw the same divide in recent cases like North Carolina Board of Dental Examiners v. Federal Trade Commission, in which Roberts and Justice Kennedy recognized that an entity controlled by financially interested market participants was functionally a private actor, whereas Scalia, Thomas, and Alito were happy to deem them public actors as long as the state labeled them a state agency. If you want to divide the conservative Justices, at least when it comes to economic regulation, the best strategy is to pay attention to this fault line between the economic and formalist conservatives.

Second, we are finally going to learn just how well Obamacare works. So far it has worked surprisingly well given that it has featured a ridiculously long phase-in and a series of legal minefields. The one thing insurers hate most is uncertainty that cannot be actuarially estimated. With the last of the major legal minefields cleared away, we can expect to see all insurers more aggressively signing up the uninsured with lower premiums now than they could otherwise have charged. I was initially opposed to the Act on two grounds: (1) the mandate seemed likely to drag the Act into needless political controversy; and (2) I didn’t think the Act did enough to make health care more efficient. I think I was right on the first point, but have been surprised at how well Obamacare’s unsung regulations on health care delivery have done in reducing cost inflation. With this Supreme Court decision in place, I think that Obamacare will succeed even more in the future, as more of the population is combined into a common insurance pool that can be more efficiently regulated. Perhaps with the basic framework settled, both parties can even unite on bipartisan efforts to improve the efficiency of the system.

Third, the opinion confirmed an important limit on the Chevron doctrine, which determines when courts defer to agency interpretations of ambiguous statutes. Much as Chevronistas hate to admit it, one of the many exceptions to Chevron is the “extraordinary case” exception that applies to questions of deep “economic and political significance.” Sheer importance is not enough because Chevron itself deferred to agency regulations on an extraordinarily important environmental policy that affected all industrial manufacturing. The decisive “political significance” that is also required is a clear political conflict between the agency/presidential view and what the current Congress would be willing to enact.   As I pointed out before the opinion, Chevron “deference applies only when the agency action is a good proxy for current enactable preferences,” which meant it did not apply here because “the IRS interpretation is clearly contrary to the views of the current Congress.” The King decision further entrenches this important limitation on Chevron.

Fourth, Justice Scalia has not lost his textualist A-game, but the textualist debate ultimately just revealed another conservative fault line. Although Roberts got the better of the textualist argument, Scalia did a remarkable job of arguing the contrary, much better than any brief on his side did. (Though alas with some unnecessary gratuitous insults hurled against the other Justices.) Roberts pointed out that the Act requires all exchanges to make health plans available to all qualified individuals and take their interests into account, but that qualified individuals would not exist in federal exchanges under Scalia’s interpretation. Scalia responded that a memo requiring all professors to take the interests of graduate students into account would make sense even though some professors may only teach undergraduates. Roberts pointed out that the Act requires all exchanges to distribute information about the amount of tax credits, including a tax credit calculator, which would not make much sense in federal exchanges where there was no tax credit. Scalia responded that federal exchanges could simply report that the tax credit was zero. Roberts pointed out that the statute said all applicable taxpayers would get a tax credit and that Scalia’s interpretation required “diving several layers down into the Tax Code” to reach the conclusion that a technical provision that defined the monthly premiums used to calculate the amount of tax credit meant that amount was zero on a federal exchange. In perhaps the most surprisingly convincing part of his opinion, Scalia stressed that the tax code often grants tax credits but creates standards that make their amount zero for many individuals and that here this makes sense because taxpayers or their dependents often move between states, making monthly calculations more suitable.

Still, I think Scalia missed two important things. The first thing Scalia missed is that Roberts only had to defend the proposition that the textual meaning was ambiguous, whereas Scalia had to show that his interpretation was unambiguously correct. Meeting one claim with an equally plausible counter-claim was not sufficient to meet Scalia’s burden, but rather it made Roberts’s point on ambiguity.

The other thing Scalia missed is that language is a purposive enterprise. Even though Scalia made a strong case that his reading did make linguistic and rational sense, the question is whether it is a plausible reading of what Congress likely wanted. Consider a case I like to use with my students. A caterer tells an employee to “collect all the ashtrays in the dining hall and put them in the van.” But when he starts to do so he realizes there are not only ashtrays on the tables but also ashtrays affixed to the walls. Should he unscrew the ashtrays from the walls and take them to the van? The plain meaning surely says so because the text says “all ashtrays.” This reading also could be rational: one could well imagine the catering contract also covered replacing the affixed ashtrays or cleaning them offsite. But this reading is an Amelia Bedelia-like reading that ignores what any real caterer likely wants.

Roberts and Kennedy are the sort of textualists who instead read text to advance likely purpose. Even though it is not linguistically nonsensical to impose on all exchanges obligations toward qualified individuals and to report the amount of tax credits, even though no qualified individuals or tax credits would exist in federal exchanges, it is implausible that any real legislature would want to do so. And even though one might for technical reasons want to measure tax credits relative to monthly premiums that could vary with individual movements across states, it would be odd to have such a technical provision be the only announcement of an extreme sanction like removing all tax credits from states that refused to set up their own exchanges. Legislatures do not, as Scalia, recognizes elsewhere, “hide elephants in mouseholes.”

We are used to dividing interpretation methodologies between purposivists (who do things like read legislative history) and textualists (who regard legislative history as a little more distasteful than pornography). But I think we need to recognize that there is another divide among the textualists: between the purposive-driven textualists like Roberts and Kennedy, who focus on text but understand that language is a purposivist enterprise, and abstract textualists like Scalia, Thomas, and Alito, who prefer to interpret text in a purely abstract way free from any consideration of purpose or functionality, out of a fear that any consideration of legislative purpose or functional desirability will morph into the imposition of judicial views.

Finally, it was to me remarkable that Scalia, Thomas, and Alito were willing to so blithely say it would be perfectly sensible to interpret Obamacare to threaten to withhold tax credits in order to induce states to create exchanges. As I have shown elsewhere, that position conflicts with their earlier conclusion that it was unconstitutionally coercive for Obamacare to threaten to withhold Medicaid funding to induce states to agree to a Medicaid Expansion. Kennedy’s vote with Roberts in this case I think reflects not only the fact that he has always been more of a purposive-driven textualist, but also that Kennedy (as he indicated at oral argument) takes that doctrine of unconstitutional coercion seriously.