King v. Burwell, NFIB v. Sebelius, and the Coercive Spending Power

November 12th, 2014

Under the petitioner’s reading of the ACA, states were presented with a choice: (a) establish an exchange, so your residents can receive tax subsidies, or (b) don’t establish an exchange, thereby denying your residents tax subsidies. The former was the carrot, the latter was the stick. In the absence of subsidies, insurance would become unaffordable, millions would drop out of the market, and the insurance markets would go into a tailspin. Let’s assume for a moment this reading is correct (and for purposes of my post, a majority of the Justices would have to agree for this argument to make any sense).

If this is the right reading of the statute, is there really a meanigful choice? Your options are either (a) establish an exchange or (b) make insurance in your states unaffordable, sending your market into an adverse selection death spiral. Under the Medicaid Expansion opinion in NFIB v. Sebelius, would this choice be coercive under the spending power? I think the argument follows from NFIB, though it would require some extension of the doctrine.

Brian Beutler (not Jeff Rosen) sums up the issue:

Because if that’s how you read the statute, then you’re positing that Congress has used its spending power to compel states to set up their own exchanges. But if it’s unconstitutionally coercive for the federal government to say “take this [i.e. Medicaid expansion money], or lose all of your existing federal Medicaid funds,” it might also be unconstitutional for the federal government to say, “do this [i.e. set up an exchange] or we’ll break the insurance markets in your state.”

Rosen notes that several states, that have established exchanges, have argued that the Petitioner’s reading of the ACA would be unconstitutional:

In a similar spirit, several states, including swing states like Virginia and North Carolina, along with conservative redoubts like Arkansas and Mississippi, have argued in an Amicus brief that the challengers’ reading of the statute renders the subsidy scheme unconstitutional because Congress can’t use its spending power to penalize states without a heads up. “Congress must give States clear notice of conditions imposed under Spending Clause statutes,” the brief reads. “But there was no such clear notice here.”

The states seem to use this argument to reject the idea that Congress could have designed the statute to set up a spending clause violation. But this neglects the fact that the Medicaid expansion worked in a very similar function–accept the expansion or stand to lose all of your funding. As I explain in this piece, the Medicaid expansion of Obamacare was designed to punish uncooperative states. Granted, there were some differences between old Medicaid and new Medicaid. But the general idea is the same. Join Obamacare or throw your state into turmoil. Congress was not concerned about the constitutional violation for Medicaid, so why would they be concerned about it for the subsidies. (They weren’t really concerned about any constitutional issues, but see Unprecedented for more details on that).

This argument may be more helpful to petitioner than Respondents. The presence of this issue, which is lurking in the background, could provide an alternate route to victory for King. Follow me here. If there were five votes to find a violation of the Spending Clause, the entire provision would be invalidated. Regardless of whether a state builds an exchange, or not, subsidies cannot be condition on those grounds. Congress could have passed a statute giving subsidies to all states, but they did not. In the absence of a new statute, no subsidies could be paid out to any state residents. That is effectively what the joint dissent did in NFIB–invalidate the Medicaid expansion, even for those states that wanted to join. Texas and California would be equally exempt.

But the Chief Justice, along with Justices Breyer and Kagan, did something else. Rather than invalidating the Medicaid expansion as applied to all states, they split the difference. States could choose to opt in would. States that did not want to opt in would not. They chose a narrow approach that gave states an option. Texas could stay out, but California could join.

The spending clause option presents similar dynamics. If there are five votes who find that the choice between setting up an exchange or going into a death spiral is coercive, that would be a broad judgment that affects all states. But, a more narrow option remains. Simply invalidating the IRS rule (the relief King seeks) leaves the statute in place, and those states that have established an exchange would still be able to receive subsidies. Texas would not be bound, but Californians would receive subsidies.

Jeff Rosen alludes to this, calling it a JGR middle finger:

But if the Court’s reading of the subsidy scheme renders the statute unconstitutionally coercive or injurious, then, per U.S. v Booker, the Court should ask “what ‘Congress would have intended’ in light of the Court’s constitutional holding.” And in so doing, Roberts et al would probably have to make the establishment of an exchange optional, with no penalty for using the federal fallback, just like they did when they made the Medicaid expansion optional. Along similar lines, Roberts could find for the government by invoking the principle that the Court ought to interpret the statute in a plainly constitutional way, rather than trespass into difficult constitutional questions.

The beauty of this scenario is that in addition to being a logically and intellectually consistent conclusion for Roberts to reach, it would also raise an elegant middle finger to all of the people on the right who called his integrity into question after he saved the ACA. I wouldn’t bet on this outcome. But let’s hope he doesn’t read this article anyhow.

I don’t think this answer is crazy. If there are five votes to expand NFIB, and find a spending clause violation, the Court could act moderate and only invalidate the rule. And in truth, striking down the rule here will not serve as much of a precedent for any future cases. But, for the Chief to lay down some spending clause dicta may be keeping with his commerce clause dicta in NFIB. Maybe we’ll get another 4-1-4.

This term, and my next book, is shaping up to be more interesting every day.

Update: I dashed this post off before running to catch a flight, and I missed an important point which makes most of this argument incorrect. If the Court invalidates the IRS rule, then all we are left with is the coercive statute. In other words, the IRS rule obviates the coercion problem. I’ll leave the post up as is, but please read it with this caveat.