Thursday, August 7, 2014
In an earlier post I showed how to analyze the Obamacare state subsidies of the Halbig decision using game theory (or, if you, like, how to analyze the earlier Health version of the bill that everybody agrees would have ruled out federal subsidies for 4 or 6 years--- I’ve seen both numbers). Game theory thinking led me to another angle on the case.
The Halbig case, with its issue of what it means for an insurance exchange to be “established by a state,” brings up the general question of how to decide whether a phrase in a statute is enough of a mistake that a judge should take action and change it. Sometimes it is easy--- if the bill says that “3 mtrillion dollars” shall be spent on a bridge, everyone would agree that it was meant to be 3 million dollars, even though it sounds more like 3 trillion.
Here is one test. The judge should ask himself what the legislature would have done if someone had pointed out the phrase and asked them to re-vote on it as it was worded.If they’d still vote for it, good, the challenge fails. If they’d vote against it, then some thought needs to go into figuring out what they would have voted for instead, whether no bill at all would have passed, or whether it’s just too hard for the court to figure out what they would have voted for. Applied to the “3 mtrillion” bill, clearly
(a) nobody in the legislature at any time would want to spend 3 trillion dollars on a bridge, so the bill so worded would fail, and
(b) if the legislature at the time had been presented with “3 million” they would have voted for it.
The test needs refining though, because “legislature” is ambiguous. Legislatures have separate chambers and their composition changes over time, and a single bill requires lots of votes on different drafts, procedures, amendments, and so forth. So we would have to ask about the legislature at a particular point in time.
Every bill has to pass through a number of what political scientists call “veto gates”--- decisions by different people as to whether it should proceed. Let’s skip over the early ones involving making a motion, seconding, sending to committee, voting in committee, and so forth. The big three veto gates are a vote by the House, a vote by the Senate, and signature by the President. Whatever else happens, it’s the final vote by a majority that matters, so let’s ask whether a majority of the House would have still voted for the bill if everyone had clearly understood that the meaning would be the plain language meaning (e.g., “3 trillion dollars”), whether the Senate would still have voted for it, and whether the President would still have signed it. In the bridge example, it’s easy. The House would have voted against it, and someone would have proposed a “3 million dollar” substitute bill. If the House had passed the 3-trillion-dollar bill, the Senate would have voted it down, and proposed the substitute. If both had passed it, the President would have vetoed it and asked Congress to come up with a substitute.
Now let’s apply this test to Halbig. The last veto-gate was whether President Obama would sign the Obamacare bill with no subsidies for federal exchanges. He clearly would have; he’d have preferred a different bill, but for the President it’s take it or leave it. If he vetoed it, it was clear no substitute would get through the House and Senate.
How about the Senate? They were presented with a bill that originated in the House (to avoid the Origination Clause problem being litigated in Sissell, that revenue bills must originate in the House), though its language had been passed by the Senate earlier, so it wouldn’t have to go through the reconciliation process unless it was changed. They wouldn’t have wanted to vote No, because then they’d have to propose a substitute, and no substitute would pass the House and the Senate because the new senator from Massachusetts would vote against it. So they’d have voted Yes, even fully understanding that the bill wouldn’t allow subsidies for federal exchanges.
How about the House? If they’d voted No they could have proposed a substitute bill. But they knew that any substitute bill couldn’t pass the Senate. So the House would have voted Yes, even fully understanding that the bill wouldn’t allow subsidies for federal exchanges.
This is all rather involved, with details that are special to Halbig and will vary from case to case. The general method answers definite and relevant questions about intent, however. It handles the problem of what it means for a *group* of people to have an intent that I posted on earlier by asking how a majority of them at a particular time in history would vote, the particular time that the Constitutions says is the only one that matters. Its weakness lies in that it assumes that we should ask how someone should vote on a clear version of the bill. This seems innocuous, but actually sometimes bills are purposely written to be ambiguous or to contain boobytrap clauses so as to allow compromises and increase the number of votes for passage. How to deal with that I don’t know. I’d start by reading “Promoting public-regarding legislation through statutory interpretation: an interest group model,” Jonathan R. Macey - Columbia Law Review, 1986.
This discussion illustrates two other things. First, it is an addition to the heap of observations on the mess that is the concept of legislative intent--- though as the 3-trillion-dollar bridge example illustrates, we’ve got to do something with it in certain cases. Second, it is a reminder that the Obamacare bill only passed through a lot of procedural artfulness. That means “intent” is even more dubious a concept than usual, and so courts should try to avoid it in this case more than usual. The House Democrats couldn’t pass the bill they wanted n 2009, and the Senate Democrats couldn’t pass the bill they wanted in 2010, so they ended up with something neither liked.