Tuesday, July 22, 2014

A Game Tree for Harry Reid vs. Indiana and the Halbig Obamacare Case


I just posted on the DC Circuit’s 2-1 decision against the Government in Halbig, the Obamacare case about whether “an Exchange established by the State under section 1311” means an exchange established by the state, so that in locations where there is no state exchange, the individual mandate and subsidy and employer mandate won’t apply.

Nobody really could believe that the language of the statute is ambiguous, so though the Administration does help lighten our day with some humorous arguments to that effect, their main arguments are about intent. Congress could not possibly have meant to incorporate provisions in a statute that would thwart the intent of those who voted for it! Well, that’s humorous too, since it’s not uncommon for politicians to shoot themselves in the foot. Indeed, given Obama’s current ratings in the polls, one might argue that the Courts should reverse the result of the 2012 election; voters couldn’t possibly have meant to re-elect Obama, and no doubt a majority of them would now admit they were mistaken. But let’s move to where law-and-economics can be helpful to the DC Circuit when it takes up Halbig en banc, as everyone thinks it will unless the case goes straight to the Supreme Court.

In his dissent, Judge Edwards says:

[N]o legitimate method of statutory interpretation ascribes to Congress the aim of tearing down the very thing it attempted to construct…

Appellants in this litigation have invented a narrative to explain why Congress would want health insurance markets to fail in States that did not elect to create their own Exchanges. Congress, they assert, made the subsidies conditional in order to incentivize the States to create their own exchanges…

The simple truth is that Appellants’ incentive story is a fiction, a post hoc narrative concocted to provide a colorable explanation for the otherwise risible notion that Congress would have wanted insurance markets to collapse in States that elected not to create their own Exchanges….

The majority thinks it unremarkable that Congress would condemn insurance markets in States with federally-created Exchanges to an adverse-selection death spiral.

      A little bit of game theory goes a long way. Here’s why Congress would indeed condemn insurance markets in States with federally-created Exchanges to an adverse-selection death spiral. It has to do with payoffs on and off the equilibrium path of a game. Or, since I’m not trying to be pedantic, with why people make threats.



     Let’s use a story. Suppose Harry Reid has a choice between passing a bill saying “No state exchange, no subsidy” or a bill saying “Fed exchange OK for subsidy.” He would prefer the states pay for the exchanges, since even a small saving like that will help reduce the cost to the federal government. He’s got a majority in the Senate, so he can pass whichever bill he wants. His payoff is 10, let us say, if the federal government has to raise taxes to pay for the exchanges, and 12 if the states pay. For the states, let’s call the player “Indiana”, since it’s one of the conservative states from which Reid expects trouble. Indiana will get a payoff of 0 if the federal government pays for the exchanges, and -1 if it has to pay for the exchanges itself, Reid thinks. Reid calculates, though, that under the “No state exchange, no subsidy” bill, Indiana’s payoff will be -3, because it will lose the subsidies and that’s more important than the expense of the exchanges. To be sure, Reid’s payoff would be -20 under that bill if Indiana didn’t establish an exchange, but that would be for Indiana to shoot itself in the foot, which politicians never do. Thus, Harry Reid passes the No State Exchange No Subsidy bill, Indiana establishes an exchange, and Harry gets his maximum possible payoff of 12. Harry has refuted Judge Edwards: he has purposely passed a bill that would cause chaos in Indiana and wreck Reid’s own policy--- but only if Indiana makes a mistake.

    Unfortunately, it was Harry who made the mistake. Indiana’s payoff from establishing the state exchange is -4, not -1. Harry has forgotten about the issue of principle: the Indiana conservative statehouse hates government health care even more than they love subsidies. And so Harry ends up with a payoff of -20.

     The game could be expanded realistically. At the cost of some complexity, we could formally model Harry’s uncertainty over whether Indiana’s payoff was -1 or -4, making this into “a game of incomplete information.” That would just be a technical cleanup, though--- the game would behave much the same. Or, we could also add another player: the Supreme Court. Suppose Harry thinks there is some chance Indiana’s payoff is actually -4, so his No Subsidy bill would backfire. He also knows, however, that some judges, like Judge Edwards, would like to come to his rescue. Thus, we might add a move by the Supreme Court at the end of Refuse to Establish State Exchange. The Supreme Court could keep the law as is, or change it to “Fed Exchange OK for Subsidy”--- in which case Reid is back to his +10 payoff. Or, even better, the Supreme Court might choose the move, “Force Indiana to set up an exchange no matter what the statute says,” and Harry could get his +12 payoff no matter how Indiana moved. I’ll leave that game as an exercise, or maybe for an amicus brief (anybody interested in writing one on this point?)

     This idea is familiar in law and economics in the form of the idea of “penalty defaults” in contract law. Professors Ayres and Gertner pointed out in 1989 that one way a judge could respond to sloppy contracts in which the parties made some clause ambiguous to try to get the courts to go to the trouble to sort it out would be for the court to pick a purposely harmful default clause to use if the parties tried that tactic. For example, if it was unclear which party would receive interest on an escrow account in a merger deal, the courts might want to say in advance that all the interest goes to the Taliban, so everybody loses. That would make them take more care with their drafting. A very recent example in another context is the case of the conservative professor who sued the U. of Iowa law school for political bias in hiring. The jury deadlocked and was dismissed, but somehow (I forget details) the judge called them back two minutes later and they were in agreement. The appellate court reluctantly said there’d have to be another trial, because though in this case it was satisfied that the jurors hadn’t had time to confer with outsiders and a new trial would be something of a waste, it was better to have a bright line rule.

Ian Ayres & Robert Gertner, Filling Gaps in Incomplete Contracts: An Economic Theory of Default Rules, 99 YALE L.J. 87 (1989).

Ian Ayres, Ya-Huh: There Are and Should Be Penalty Defaults, 33 FLA. ST. U. L. REV. 589, 597 (2006).



| Permalink


Post a comment