Halbig, Statutory Interpretation, and Lessons I Learned in Practice

Reading all the interweb debates in the wake of Halbig and King, I’ve grown increasingly frustrated. People seem to be talking past each other—perhaps for the reasons Will Baude outlines here. Arguments about statutory text are rebutted with arguments about congressional intent, which is often gleaned from press coverage during the ACA’s legislative debates.

But I think it cuts deeper than that. The nature of the post-Halbig debate feeds into the core stereotypes and motive-impugning assumptions that legal conservatives and liberals have about each other. Conservatives willingly play the role of hyper-technical textual literalists with simple-sounding (and superficially appealing!) arguments tailor-made for Twitter: “state” means “state”; it doesn’t mean “federal.”[1] Liberals, on the other hand, ease naturally into the opposite stereotype, countering this plain-language reading of the statute with arguments that could be caricatured as: “I was reading The New York Times daily coverage of the ACA debates and they didn’t mention any of this, so it can’t be what Congress actually meant.”[2]

I’ve come away with two observations, both of them focused on the liberal side of the argument. The first is that we liberals should listen to my former boss Susan Brune. The second is that if we want to preach to anyone other than the converted, we need to focus on the ethos of our rhetoric in the arena of statutory interpretation.[3]

But first, Susan Brune. After clerking, my first job as a practicing lawyer was with Brune & Richard in New York City. (It was a great firm; I miss working and not working with my former colleagues there.) Early on, I was drafting some sort of persuasive brief for Susan, and I included some standard line you see in a lot of briefs, something like this:

The government provides no evidence linking Mr. Defendant to the transaction here at issue.

She hated it. Why? Because I said “no evidence.” That was wrong for two reasons. First, to win the motion we didn’t need to establish that there was no evidence against our client; we just needed to establish that the evidence was insufficient under whatever the relevant standard was. Second, there is always evidence against your client—it might be weak, it might be circumstantial, it might be based on a chain of inferences, but it’s always there.

When you say “no evidence” you’ve therefore set yourself up for a disaster. The other side is surely going to respond with something like this:

The Defendant claims there is “no evidence” tying him to the transaction. But the record is replete with evidence connecting him. First, the Defendant’s then-girlfriend had access to the relevant account . . . .

And so on and so on. You then wind up fighting about picayune details like your client’s girlfriend’s access to the account or whatever little specks the other side digs up. And because you started with a “no evidence” claim now you’re on the defensive. You’ve lost credibility. You told the Judge that you should win because there was no evidence. But now there is some (weak) evidence! Does that mean you should lose? Ugh.

The liberal critics of the Halbig decision have fallen into this trap not once, but twice. First, folks started by saying that the decision (or the challengers’ arguments) were “absurd!” In matters of statutory interpretation, “absurd” has a special meaning. A law isn’t absurd just because you don’t think that’s what the legislature intended. The law must be objectively absurd! For example, a statute that makes it a crime to block the delivery of the mail should not be applied to a sheriff who arrests an on-duty mailman.[4] In other words, there’s just no way any Congress could ever possibly mean to literally apply the statute in this way!

The claim of absurdity in Halbig was easily rebutted by conservative scholars who pointed out that in numerous other instances, Congress had done precisely what the critics were calling “absurd” in this case: Congress had conditioned benefits for a state’s citizens on the state’s agreement to go along with some sort of congressional plan. If Congress had set up precisely this sort of system of sticks and carrots before, then it can’t be “absurd” to say that’s what Congress was doing here.

Of course, liberals had a reply: But that’s not what Congress meant to do this time. At this point, however, we are already down the rabbit hole, arguing about whether something is really absurd and whether it really was so ridiculous to think Congress might have wanted to set up such a system here. That argument, however, is a far cry from the argument liberals should be having: What on earth does the statute actually say and how does it work in context? Just like the “no evidence” claim, we’re now off track and we’re on the defensive.

A similar thing happened with arguments about congressional intent. Yes, the vast majority of contemporaneous analysis and coverage of the ACA drafting and debates assumed that state citizens enrolled under federal exchanges would be eligible for tax subsidies under the ACA. Nicholas Bagley has a nice summary here. But that’s not really the argument liberals were making in the wake of the Halbig decision. Nope. Instead, what I heard was a lot of “no evidence” claims—a lot of articles along the lines of “Nobody with any knowledge of the ACA who was paying any attention or was in any way involved during the drafting of this legislation ever even considered that the subsidies wouldn’t apply to citizens of states with federal exchanges.”

You know what happened next. Folks found two clips of a person intimately involved with the legislation saying that exact thing. And now, because liberals led with the “no evidence” sort of argument, they are left on the defensive while conservatives think they’ve found their silver bullet. Of course the Gruber clips need to be weighed against the greater context, but the “no evidence” argument has been lost. It wasn’t even an argument we should have had in the first place.

So what sorts of arguments should we be making? If we want to actually convince people (and we’ll need to do that as long as there are Republican Presidents and Senators) we should focus a bit on the ethos of our statutory arguments. In rhetoric, you need to have logic on your side, but you also need to connect with your reader, you need to make appeals to credibility, community, virtue, wisdom, goodwill, etc. You need to reach out to the other person and convince them that you actually have something worth listening too.

Far too often (at least in my opinion) the anti-Halbig arguments treat the statutory text like a stumbling block, as if the goal were simply to make a plausible case for ambiguity so the problematic text can be side-stepped. And then it’s off into the comfortable land of intent and purpose. That’s not to say the arguments wholly ignore the text, or that there is no real engagement with the text. But the ethos of these arguments drives toward the author’s understanding of the general statutory purpose, with the text just a hurdle in the way.

Here are a couple of examples, but really, they are meant to be generally illustrative:

In an article for the National Law Journal, Erwin Chemerinsky and Samuel Kleiner begin as follows:

Conservatives fiercely oppose the Patient Protection and Affordable Care Act, and so it is not surprising that two conservative judges on the U.S. Court of Appeals for the D.C. Circuit handed down a decision that would gut the law.

That’s certainly not going to gain you any converts. But how does the legal analysis start?

The Affordable Care Act was intended to ensure that almost all Americans have health care coverage.

No text. No statutory language. Just right to congressional intent. And not even intent with respect to the meaning of the phrase, but a general overarching purpose. The ethos of this argument is loud and clear: “Text schmext. I know what the law really means.”

And here’s Andrew Koppelman in the New Republic, after setting out the Halbig plaintiffs’ general claim:

This reading of the statute makes no sense. Courts don’t read provisions in isolation. Here are a couple of boring, standard rules of statutory interpretation: don’t take phrases out of context; try to make sense of the law as a whole. The ACA’s purpose was universal coverage.

The relevant “context,” apparently, is the general purpose of the statute, not the text of related provisions. If that’s the case, then why not interpret “carrying out 1 or more responsibilities of an Exchange” to mean “YOU MUST ALL BUY HEALTH INSURANCE OR YOU WILL BE IMPRISONED!!!”

I kid, I kid. The thing is, I agree with most of the arguments made in the articles I just quoted; I just don’t think they are the main arguments we should be making. Unless the intended responses are (1) people we already agree with patting us on the back or (2) people we disagree with howling that we’re ignoring the statutory text in a fit of liberal-elites-know-best-ism, then we need to start where all statutory interpretation theory says we should start: with the text of the statute.

And once we start there, we need to stay there for a while. Statutory text is not a weigh station. Almost all forays into statutory interpretation start with some statement along the following lines: “We begin, as we must, we the text of the statute.” Well of course we do. But for those who fear that the text is a flimsy restraint on judges seeking to enact their own policy preferences, the pronouncement to begin with the text has all the limiting power of construction-paper handcuffs. Imagine hearing a wedding vow that said: “Of course, I will begin by being faithful to you.” Yikes.

So to earn some credibility, that ethos, we really need to wrestle with the statutory text a bit. Nicholas Bagley has done a very nice job of this, and others have too. But if we want the discourse of statutory interpretation to move beyond mudslinging, then from a rhetorical perspective we need to view the text as more than just a box to check.

With that said (and goodness, if anyone is still reading) here are a few text-based questions for folks who think this case, as a textual matter, is really just as easy as “state ≠ federal” in the ACA. These are not supposed to be “gotcha” questions; there likely may be easy answers. But I think anyone who says the ACA unambiguously means only state-established exchanges when it uses the phrase “an Exchange established by the State” should have simple and unambiguous answers.

1. If the term “Exchange” is defined in the statute as “an American Health Benefit Exchange established under section 1311 of the [PPACA],” PPACA § 1562(b) (emphasis added), and if section 1311 only provides for state exchanges (with section 1321 providing for federal exchanges), then does the term “Exchange” standing alone, without modification, mean only state-established exchanges?
2. Relatedly, if an Exchange is defined as something established under section 1311, how can the Secretary ever create an Exchange as that term is used in the rest of the PPACA?
3.  If the term “Exchange” is meant to refer inclusively to federal- and state-established exchanges, then what is the effect of the “established under section 1311” language in section 1562(b)?
4. If the PPACA sets up a two-tiered system for federal- and state-established exchanges, where are the provisions governing exchanges “established under section 1321” or “established by the Secretary under section 1321”? If they are absent, why are they absent?
5.  If the term “Exchange” is meant to refer generically to federal- and state-established exchanges, with the phrase “Exchange established by the State” reserved solely for state-established exchanges, then why would the PPACA make federal-established exchanges subject to annual audits by the Secretary if the Secretary is the one operating the exchange?
6. Is a regional exchange operated by a collective of states an “Exchange established by the State under section 1311”? What if one state “established” the Exchange but the other state simply permits the operation of the exchange within its borders, pursuant to section 1311(f)(1)?
7. What is the difference between an Exchange, an “Exchange established pursuant to this title,” section 1303(a)(1)(D), an “Exchange established under this Act,” section 1312(d)(3)(D)(i), an “Exchange established under section 1311,” see, e.g., section 1331(e)(2), and an “Exchange established by the State under 1311,” section 1401(b)(2)(A)? The PPACA uses all of these differing phrases to refer to exchanges.
8. Relatedly, could a state establish an exchange pursuant to a section other than section 1311? Could the Secretary establish an exchange pursuant to section 1311? If not, then why say “established by the State pursuant to section 1311”?

On October 30, I was invited to participate in a Cato Institute panel discussion on Halbig, King, and related challenges to the Affordable Care Act. In the wake of the conference, I’ve written a few posts based on the panel discussion. If you’d like to watch the entire law-focused panel, I’ve created a C-SPAN clip here. My thoughts based on the conference are at Part I (Isolationism), Part II (Textualism), Part III (The Whole-Text Canon), Part IV (Halbig’s “Two Exchanges” Problem), and Part V (Creeping Constitutionalism).

[1] For some reason, this simplistic argument has a sort of “It’s Adam and Eve, not Adam and Steve” ridiculousness to it, both in its pith and its irrelevance.

[2] If you read this and you don’t think it applies to whatever article you’re thinking about, then just assume I’m not talking about that article.

[3] I’m terrible at writing catchy ledes that make people want to keep reading. And I really need an editor. Sorry.

[4] Thanks to Veronica M. Doughterty, Absurdity and the Limits of Literalism: Defining the Absurd Result Principle in Statutory Interpretation, 44 Am. U.L. Rev. 127, 137-38 (1994), for providing me with this example case: Kirby v. United States, 74 U.S. 482 (1868).