Law & the Courts

The Supreme Court Proves It Didn’t Mean What It Said in King v. Burwell

Statue outside the Supreme Court in Washington, D.C. (Jonathan Ernst/Reuters)
In a new decision, the justices have embraced the same logic they rejected in King.

In June 2015, the Supreme Court ruled that Obamacare provided subsidies to buyers of health insurance on the federally operated exchanges, not just exchanges established and operated by a state. Many legal observers at the time, myself included, argued that the decision in King v. Burwell was a politically driven outcome that disregarded longstanding rules for how the Court reads statutes, and that the Court in the future would have to either accept a dramatic sea change in those rules or admit (at least implicitly) that King v. Burwell was a political, not a legal, decision. Well, what do you know? The Court’s decision last month in Digital Realty Trust, Inc. v. Somers makes it crystal clear that the Court does not take King v. Burwell seriously as a legal precedent, and would have decided that case differently if it had not been so politically charged.

To recap, the Court in King v. Burwell upheld a rule passed by Obama’s IRS that extended subsidies to buyers on the federal exchanges. To reach that conclusion, however, the Court had to leapfrog the language of the statute, which made its meaning obvious in four ways (all emphasis added):

  1. It tied subsidies to policies purchased on “an Exchange established by the State.”
  2. It defined “State” as “each of the 50 States and the District of Columbia,” rather than the federal government.
  3. It further limited the definition of an “Exchange established by the State” to one established “under Section 1311” of the ACA (42 U.S.C. §18031), a provision that dealt with state-established exchanges, while a different section (42 U.S.C. §18041), which was not referenced in the subsidy provision, dealt with federally established exchanges.
  4. It used the phrase “established by the State” elsewhere in ways that could not be construed to refer to the federal government.

Hanging a gigantic red flag on this evasion of the statutory language, the IRS rule said that subsidies would be available “regardless of whether the Exchange is established and operated by a State . . . or by HHS” (emphasis added) — in other words, the rule specifically said it would apply regardless of the very terms used in the law. The IRS was originally uncomfortable doing this; a subsequent congressional investigation revealed that IRS and Treasury officials were concerned the law had left out the federal exchanges through a “drafting oversight,” expressed concern that there was no direct statutory authority for proposing the rule, and asked HHS to come up with a basis for doing so. Eventually, though, the political imperatives of the Obama administration won out.

The Court’s decision last month in Digital Realty Trust, Inc. v. Somers makes it crystal clear that the Court does not take King v. Burwell seriously as a legal precedent, and would have decided that case differently if it had not been so politically charged.

Nobody in 2009–10 had expected the federal exchanges to be a big issue; the CBO scored the bill on the assumption that every state would establish an exchange. The challengers to the IRS rule argued that limiting the subsidies to state-run exchanges made sense as a heavy-handed incentive for states to go along – just like the same statute threatened to eliminate a state’s entire Medicaid funding if it didn’t implement Obamacare’s Medicaid expansion. Indeed, the Court in 2012 found that this was such an extreme threat that it unconstitutionally stripped the states of a free choice.

But by 2015, only 16 states and D.C. were running exchanges. Many Republican-run states refused to do so, while many states that tried had to shut their exchanges down after they worked poorly. Neither group of states saw any real incentive to operate one if Uncle Sam would pick up the slack. Faced with the prospect of unraveling the federal exchanges in the other 34 states, Chief Justice Roberts (who wrote the 6–3 opinion in King v. Burwell) and Justice Kennedy sided with the Court’s four Democratic appointees to uphold the IRS rule, lamely protesting that “a statutory term may mean different things in different places” (even though the statute in question included an explicit definition!) and focusing mainly on the results-driven argument that the challengers’ interpretation . . .

. . . would destabilize the individual insurance market in any State with a Federal Exchange, and likely create the very “death spirals” that Congress designed the Act to avoid. . . . Without the tax credits, the coverage requirement would apply to fewer individuals. And it would be a lot fewer. . . . Those credits are necessary for the Federal Exchanges to function like their State Exchange counterparts, and to avoid the type of calamitous result that Congress plainly meant to avoid. . . . 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.

To drive its point about the statutory “purpose” of the law home, the Court cited 2014 statistics about the size of the federal compared to state exchanges (87 percent of subsidy recipients were through the federal exchange) that nobody could have anticipated when the law was written five years earlier. Though Roberts, while ignoring many of the more specific statutory-language points raised by Justice Scalia in dissent, conceded that the challengers’ “arguments about the plain meaning of [the section on subsidies] are strong,” he concluded that “in this instance, the context and structure of the Act compel us to depart from what would otherwise be the most natural reading of the pertinent statutory phrase.”

But Supreme Court decisions about how to read laws aren’t supposed to be only for “in this instance.” Enter Digital Realty, a case involving the sometimes controversial but not so politically fraught question of how to protect “whistleblowers” who report corporate fraud. The 2002 Sarbanes-Oxley Act, passed in the aftermath of the Enron and WorldCom scandals, allowed employees to sue for retaliation by their employers if they had provided certain types of information or assistance to the Securities and Exchange Commission, the Justice Department, Congress, a number of other agencies, or internally to a “person with supervisory authority over the employee.” The 2010 Dodd-Frank Act, revisiting the subject after the 2008 financial crisis, added to the protection of whistleblowers in two ways: It allowed retaliation lawsuits without many of Sarbanes-Oxley’s procedural restrictions (such as having to first file a complaint with the Labor Department) and it offered monetary bounties to encourage whistleblowing that resulted in a successful SEC lawsuit or enforcement proceeding.

The question in Digital Realty was whether Dodd-Frank’s anti-retaliation protections for whistleblowers protected only whistleblowers who went to the SEC, or covered everyone who reported misconduct to the same list as in Sarbanes-Oxley: prosecutors, Congress, their boss. Like Obamacare’s (a statute written by the same Congress), the language of Dodd-Frank ought to be pretty clear: the new section of the statute (which covered both anti-retaliation and bounties) defines a “whistleblower” — in a definition that “shall apply” for the whole section — as “any individual who provides . . . information relating to a violation of the securities laws to the Commission, in a manner established, by rule or regulation, by the Commission” (emphasis added). It goes on to protect whistleblowers against retaliation for three categories of whistleblowing (all emphasis added):

  1. “providing information to the Commission.
  2. Assisting or testifying in any investigation or proceeding “of the Commission.”
  3. “Making disclosures that are required or protected under” Sarbanes-Oxley.

The last category is what caused the dispute at issue in Digital Realty. Because Dodd-Frank protects the same disclosures to your boss that are protected under Sarbanes-Oxley, it was argued, it must cover all the people who make those disclosures even if they never went to the SEC, or didn’t do so until after they were fired.

The apparently clear statutory definitions led Obama’s SEC to propose a rule treating “whistleblowers” the way the statute defined them, and for good measure to say it “tracks the statutory definition of a ‘whistleblower'”: just the people who go to the SEC. But then, as Obama’s IRS did in the case of the subsidy rule, the SEC reversed itself. It issued a final rule that used the statutory definition of a whistleblower in the bounties provision, but defined the term more broadly in the anti-retaliation provision, saying you could count as a whistleblower “whether or not you satisfy the requirements, procedures and conditions to qualify for an award.” Once again, the Obama administration was making rules the way it wanted, “whether or not” the law passed by Congress was actually written that way.

The U.S. Court of Appeals for the Ninth Circuit agreed that the SEC could rewrite the statute, relying heavily on King v. Burwell:

Terms can have different operative consequences in different contexts. See King v. Burwell . . . . The use of a term in one part of a statute “may mean [a] different thing” in a different part, depending on context. . . . This is true even where, as here, the statute includes a definitional provision. . . . [Dodd-Frank’s] anti-retaliation provision unambiguously and expressly protects from retaliation all those who report to the SEC and who report internally. See King. . . . Reading the use of the word “whistleblower” in the anti-retaliation provision to incorporate the earlier, narrow definition would make little practical sense and undercut congressional intent . . .

The Second Circuit had taken a similar tack. Judge Owens dissented from the Ninth Circuit’s decision, wryly chiding the majority for its reliance on King v. Burwell:

Both the majority here and the Second Circuit in Berman rely in part on King v. Burwell . . . to read the relevant statutes in favor of the government’s position. In my view, we should quarantine King and its potentially dangerous shapeshifting nature to the specific facts of that case to avoid jurisprudential disruption on a cellular level. Cf. John Carpenter’s The Thing (Universal Pictures 1982).

When the case got to the Supreme Court, the lawyers — experienced SCOTUS practitioners with the pulse of what arguments the Court would take seriously — treated the references to King v. Burwell as a kind of embarrassment, the drunk panhandler on the subway that nobody wants to make eye contact with. The defendants’ lawyers brushed them aside as an artifact of a one-time controversy: “lower courts have placed great weight on the Court’s recent decision in King v. Burwell. . . . But this case bears no resemblance to the unique facts of King.” The employee’s lawyers cited it only once in their brief, to reference the fact that the defendants had brought it up. The Solicitor General filed an amicus curiae brief in support of the employee, and didn’t even cite it. No justice asked about it at oral argument.

Unlike the Obama administration in King, the would-be whistleblowing employee in Digital Realty didn’t get a single vote for his position. Justice Ginsburg’s opinion for the Court, which was unanimous on all but one point (about citing legislative history to show the statute’s purpose), not only didn’t follow King v. Burwell, it also didn’t even bother to cite it to explain why it wasn’t an important precedent. And along the way, the Court bulldozed its way through every argument it had accepted in King:

“When a statute includes an explicit definition, we must follow that definition,” even if it varies from a term’s ordinary meaning . . . This principle resolves the question before us.

“When Congress includes particular language in one section of a statute but omits it in another . . . this Court presumes that Congress intended a difference in meaning.” . . . Courts are not at liberty to dispense with the condition—tell the SEC—Congress imposed.

The plain-text reading of the statute undoubtedly shields fewer individuals from retaliation than the alternative proffered by [the employee] and the Solicitor General. [The employee and the SG cite statistics that 80% of whistleblowers report internally first, but] even if the number of individuals qualifying for protection under [the statutory definition] is relatively limited, “it is our function to give the statute the effect its language suggests, however modest that may be.”

Overlooked in this protest is Dodd-Frank’s core objective: to prompt reporting to the SEC . . . In view of that precise aim, it is understandable that the statute’s retaliation protections, like its financial rewards, would be reserved for employees who have done what Dodd-Frank seeks to achieve, i.e., they have placed information about unlawful activity before the Commission to aid its enforcement efforts. [All emphasis added.]

The silence of Justice Ginsburg and the rest of the Court speaks volumes. The critics were right: King v. Burwell is not a real legal precedent, it’s a political one, a dark chapter in the Court’s history not to be spoken of again.


The Latest