Harvard Study of CBO Reports Says Nothing New or Interesting about King v. Burwell

Vox's Sarah Kliff reports that Harvard University's Theda Skocpol has produced a study purporting to show Congress intended for the Patient Protection and Affordable Care Act (PPACA) to authorize health-insurance subsidies through exchanges established by the federal government—even though the statute expressly and repeatedly says those subsidies are available only "through an Exchange established by the State." Whether the PPACA authorizes those subsidies in the 36 states with federal exchanges is the question presented in King v. Burwell. The Supreme Court will hear oral arguments in King on March 4, with a ruling expected by June. Unfortunately for the administration and its supporters, Skocpol offers nothing either new or that supports the notion that Congress intended something other than what it expressly said in the statute.

What evidence does Skocpol claim to have found in support of her counter-textual interpretation of congressional intent? She combs through 68 analyses issued by the Congressional Budget Office during 2009 and 2010. She finds that in none of those reports did the CBO entertain the idea that the PPACA's exchange subsidies might be available in some states but not others. She interprets this as both "excellent evidence" and "the best objective evidence we have that no one in Congress considered premium subsidies restricted to certain states to be either possible or desirable."

Yeah, about that.

An alert Vox reader already informed Kliff that the claim that CBO never considered the possibility of exchange subsidies in some states but not others isn't exactly true. The comprehensive health care bill approved by Democrats on the Senate's Health, Education, Labor, and Pensions (HELP) Committee in 2009 (S. 1679) would have given states four years to establish exchanges themselves, after which point the federal government would establish an exchange. As my partner-in-crime-fighting Jonathan Adler and I write in an amicus brief filed with the Supreme Court in King:

S. 1679 asked each state to adopt certain health insurance regulations, and either establish an Exchange itself or ask the federal government to establish one “in” the state... S. 1679 withheld Exchange subsidies, as well as many of its insurance regulations, for up to four years until the state complied.

The CBO scored S. 1679 assuming that some states would establish exchanges early and some would not. Thus the agency's cost projections assumed that exchange subsidies would be available in some states but not in others. So we've already got a problem with Skocpol's analysis.

What Skocpol, Kliff, and her otherwise alert reader missed—which they would not have missed if they read our brief—is that the HELP bill permanently withheld exchange subsidies if a state failed to implement that bill's employer mandate. Again, from our brief:

After four years, the federal government would establish an Exchange “in” the state and implement guaranteed-issue and community-rating rules even stricter than those found in the PPACA. If a state thereafter failed to implement the bill’s employer mandate, S. 1679 withheld Exchange subsidies permanently—even in a federal Exchange.

Emphasis in original. In fact, the HELP bill would have withheld exchange subsidies in both federal exchanges and state-established exchanges if the state did not cooperate by implementing the bill's employer mandate. Even the government's amici concede the HELP bill, which was merged with the Finance Committee's health care bill to form the PPACA, conditioned exchange subsidies on states implementing the law. So much for Skocpol's claim that "no one in Congress considered premium subsidies restricted to certain states to be either possible or desirable." (Senate Democrats discarded this part of the HELP bill when crafting the PPACA, but retained and strengthened the Finance bill's language conditioning exchange subsidies on states establishing exchanges.)

Why is this relevant to Skocpol's analysis of the CBO's cost projections? Because, as Adler and I write in our brief,

the CBO likewise scored S. 1679 (the HELP bill) as providing Exchange subsidies in all states, even though—as all sides acknowledge—the bill withheld Exchange subsidies in non-compliant states.

In other words, the fact that the CBO assumed there would be subsidies in all 50 states under the HELP bill or the PPACA does not indicate that Congress did not intend to condition those subsidies on state cooperation. The CBO also assumed that the PPACA's Medicaid-expansion subsidies would be available in all states. Does that mean those subsidies were not conditional on state cooperation?

None of this is news to people who have been following King and related cases. Skocpol could have saved herself a lot of time by reading a December 2012 letter from CBO Director Douglas Elmendorf to congressional inquisitors, in which Elmendorf admits CBO staff “did not perform a separate legal analysis” of whether the PPACA only offered exchange subsidies (like Medicaid subsidies) in cooperative states. The CBO just assumed subsidies would be available in all states, under both the HELP bill and the PPACA, despite clear statutory language conditioning those subsidies on state cooperation. The simplest and kindest explanation is that the CBO shared the universal assumption among PPACA supporters—acknowledged even by the government's leading advocate in King, University of Michigan law professor Nicholas Bagley—that all states would establish exchanges or otherwise cooperate in the law's implementation. As Adler writes, "widespread state resistance was not seen as a real possibility until the dramatic GOP pick-ups in state legislatures in fall 2010."

Skocpol's study is just the latest in a series of silly and desperate attempts to bolster the government's untenable position in King v. Burwell. The law is clear, and the government is on the wrong side of it.