July 18, 2005
Two Ways of Looking at "Rational Basis"
Posted by Will Baude

David Bernstein at the Volokh Conspiracy and Christine Hurt (below) have both posted about this decision from the Wisconsin Supreme Court, striking down medical malpractice damage caps because they lack a "rational basis", and in particular, fail to limit the costs of health care. I add a few observations.

Professor Bernstein complains that the decision is "remarkably illogical":

Won't have a noticeable effect? It's possible. Is not "rationally related?" Only because the court seems to define "rationally related" as "having a guaranteed large effect."

"Rational" is, of course, a legal term of art, so Bernstein's argument is insufficient to show that the rationality threshhold is lower than the noticeable-effect threshhold.

Professor Hurt, meanwhile, brings a bit of actual empiricism to bear, and points out that there is in fact very little evidence that malpractice caps do their job. She then wonders about that term of art, "rational":
As long as the legislation has or could have an effect greater than zero, even if not noticeable? That is quite a bit of deference there. I would hope that rational basis scrutiny would require some estimation that the legislation would effect its goals to a noticeable degree.

The trouble here, which leads to some of Bernstein's and Hurt's disagreement, is that we have two rational-basis tests. For example, compare Beach Communications ("(A) statutory classification ... must be upheld against equal protection challenge if there is any reasonably conceivable state of facts that could provide a rational basis for the classification. ... In other words, a legislative choice ... may be based on rational speculation unsupported by evidence or empirical data.") with Cleburne (finding a statute to be irrational because there was no empirical evidence supporting it in the record) or Stevens' concurrence in Cleburne (suggesting that a statute is rational only if a "rational member of th(e) disadvantaged class" might support it). [The Wisconsin Court collects other cases here.]
This is why C.J. Abrahamson admits in her opinion that she is applying "rational basis with teeth," which involves some judicial scrutiny rather than sitting around like a potted plant. Since the case goes off on the Wisconsin Constitution, which rational basis to apply (toothy or toothless) is surely just a question of state law. One could object that the Wisconsin Constitution mandates some other doctrinal rule, or that there is in fact plenty of empirical evidence despite what Professor Hurt and the Court think, but there's simply no reason to think that the rational basis test imposed by a state constitution should always be the toothless one that Bernstein envisions.

Health Care | Bookmark

TrackBacks (0)

TrackBack URL for this entry:

Links to weblogs that reference Two Ways of Looking at "Rational Basis":

Comments (18)

11. Posted by Will Baude on July 19, 2005 @ 15:14 | Permalink

Maybe Article I Section 9 of the Wisconsin Constitution? "Every person is entitled to a certain remedy in the laws for all injuries, or wrongs which he may receive in his person, property, or character."

I know nothing about the provision other than what it says, but a facial reading suggests that failing to provide a person a remedy at law for certain injuries would be unconstitutional.

Incidentally, this raises the intriguing question of whether the desire to avoid raising questions under a different constitutional provision can constitute a Rational Basis (for toothless or toothed purposes). Stevens suggests maybe in his Beach Communications concurrence, but that's only for federal purposes, and only his own thoughts.

12. Posted by Ted on July 21, 2005 @ 5:02 | Permalink

Apparently, in Wisconsin, you do now.

No, because then this lineup of the Wisconsin Supreme Court will move the goalposts in a different direction—as evidenced by the fact that they had empirical evidence and chose to ignore it.

Look, if every judge were Judge Easterbrook or McConnell or me, I'd have no trouble with any level of teeth in the review, because I'd know that the court would apply the scrutiny honestly and intelligently. But no judges are selected or elected based on their understanding of public policy and economics, and as a result, it's an accident when one happens to have that understanding. Making that aspect of the job more important to the polity than the everyday exercise of the judicial power, which is what Lochnerizing does, means that the judicial branch will either be poorly situated to carry out scrutiny-with-teeth or poorly situated to handle the more mundane judicial duties.

Now, I agree with your other point that states can certainly choose to bless their courts with such disproportionate powers outside of constitutional checks and balances. But I don't think anyone in Wisconsin government except a 4-3 majority of the Wisconsin Supreme Court thinks that Wisconsin has done so.

13. Posted by Will Baude on July 21, 2005 @ 5:42 | Permalink

Well, I certainly agree that a great number of judges don't apply empirical evidence particularly well. However, I think there are two reasons not to treat that too dispositively.

First, once judges admit and make clear that they will regularly be in the habit of Lochnerizing litigators making Constitutional challenges and their amici will make more clear what the costs and benefits of their laws are. Our justice system is based, for better or worse, on a faith that the adversarial process brings out something like truth, and I don't see any particular reason that would be less true here.

Second, one shouldn't reify the legislative process. Yes, courts may get lots of things wrong, be results- rather than evidence- driven, rule outside of their areas of expertise, and so on. But all of this is true of legislators and the legislative process too. So the question is not so much "could Courts make good policy in the abstract?" but given the sausage-like way in which laws are made, would adding a level of constitutional review make things better or worse?

Anyway, I don't see why exercising equal protection review over lawsuit-damages is any more "disproportionate" or against the principle of checks/balances than exercising equal protection review over school district lines, 8th amendment review over prison medical care, or RFRA review over the federal drug statutes.

The point was not just that people in states could decide that their legislatures need second-guessing but that they should.

14. Posted by Ted on July 21, 2005 @ 13:20 | Permalink

Great judges—e.g., Easterbrook, Roberts, Scalia—take pride in noting that the question before them is not whether they think a statute or policy is wise, but whether it's constitutional.

As Epstein has noted, if the constitutional test becomes fuzzy enough that reasonable minds can differ, not only does it increase litigation expense because of increased uncertainty, but it gives judges cover to pretend that they're neutrally applying the constitution and then hide results-based reasoning behind a veneer of judicial decision-making. It's literally legislating from the bench.

When I was in law school, I hate hate hated it when attorneys would pull rank on me and dispute my theoretical musings by saying the world doesn't work that way, but I've now seen first-hand Epstein's argument come to life in far too many cases to think that an amorphous standard applied by one, three, or even nine judges will, in practice, be a workable and fairly applied standard. There's just too much incentive for judges to substitute their personal views and grab power within the system.

Perhaps people in states should decide that their legislatures need second-guessing: but the method you propose, giving that power to the judiciary with close to unfettered discretion, simply creates a third, and more elite, house of the legislature, with all of the problems associated with legislative decision-making, combined with all of the problems associated with litigation, combined with a potential loss of efficacy in the quality of judging qua judging, as judicial benches cease to be specialists in the judicial function. Why not just create that third house explicitly, and populate it with economists?

For what it's worth, I think you're not that far from the truth when you argue that "exercising equal protection review over lawsuit-damages is[n't] any more 'disproportionate' or against the principle of checks/balances than exercising equal protection review over school district lines, 8th amendment review over prison medical care, or RFRA review over the federal drug statutes," but draw the wrong conclusion from this observation.

Maybe the answer you seek is to require a super-majority of the court to overrule the legislature in such cases. That would avoid the nakedly political decisions like Ferdon, while preserving some second-guessing for cartelizing statutes. I'm not convinced that that's preferable to establishing rules that require judges to act like judges, but it's certainly preferable to a world where Ferdon is possible.

I note finally that the judicial branch, as an institution, asks for a certain level of elite treatment: their deliberations are secret, their election campaigns are supposed to be high-minded, etc. We lose the justification for that, for better or worse, and I think for worse, once the role of the judiciary is to second-guess the wisdom, rather than the legality, of the legislature's decisions.

15. Posted by Ted on July 21, 2005 @ 13:41 | Permalink

I think one point where we have a disagreement is your premise that the Ferdon majority didn't "apply empirical evidence particularly well." I'm taking the position that they didn't apply empirical evidence at all; they reached the result they wanted, and then, like trial lawyers or NFL CX-debaters, picked and chose from the evidence out there that supported that result. I don't see how anyone can read the dissents and come to a different conclusion about what happened. The majority didn't make a good-faith mistake; they knew precisely what they were doing, and there was nothing the cap proponents could've said in their briefs that would have persuaded these four justices.

16. Posted by Will Baude on July 21, 2005 @ 13:43 | Permalink

I will be one of the first to agree that all things being equal neutral rules or simple rules are better than weird complex ones in a lot of circumstances.

But when you say I "draw the wrong conclusion from th(e) observation" that the prison abuse and school desegregation cases as well as RFRA or RLUIPA cases are as complex and un-simple as Ferdon, what do you have in mind?

I thought you might be thinking per Vermeule and Thayer and others, that this means that Courts should enforce only the clear provisions of the Constitution and not the vague ones (i.e., stop 34-year-olds from running for president but leave questions like school district lines or whatever to the political branches).

But then we also have the problem of statutory law, in that a lot of statutes like RFRA, RLUIPA, Sherman &c. force judges to engage in the same kind of unpredictable review. Should judges simply abstain from enforcing statutes until they contain clearly codified instructions? Or is that different somehow?

Anyway, I take issue with your contrast between "the wisdom, rather than the legality, of the legislature's decisions." The point is that there are reasons-- I think very good reasons-- that one might sometimes, in certain limited circumstances, want to make "wisdom" or something like it a principle of legality. See, e.g., U.S. Const. Amdt. IV.

Anyway, if there is any sign that we are intending to amend the constitution and statutes to take all "fuzzy" rules out of circulation, I may sign up. But so long as we've got fuzzy civil rights review, why shouldn't states enact provisions for fuzzy economic review too? Are judges more knowledge about schools than they are about filled milk?

17. Posted by Will Baude on July 21, 2005 @ 13:50 | Permalink

In response to the second comment (#15) which I didn't see until now: Let's assume, arguendo, that four Justices of the Wisconsin supreme court did indeed ignore their judicial oaths as you suggest.

1, is there any a priori reason to suppose that elected judges are likely, on the whole, to be more dishonest and results-motivated than elected legislators?

2, is there any reason that the fact that four judges dishonestly misapplied a doctrine implies that that doctrine should never be adopted as a law?

I take it the idea is that Ferdon illustrates how easily a court can use rat. basis w. teeth to do whatever it wants. This is true enough, but I am even more pessimistic. Under any likely set of state laws, a truly dishonest state court can pretty much invent statutory and constitutional pretexts do whatever it wants, and it will not be caught or rebuked unless the *results* are unpalatable to the relevant legislative or popular bodies. So the misapplication theory seems to prove too much.

18. Posted by Ted on July 21, 2005 @ 17:22 | Permalink

Speaking of complexity, the debate is rapidly spinning into something more complex than I can give fair attention to; I apologize in advance then if you find my responses lacking. I'll also concede in advance that Will has almost certainly done more and deeper thinking about theoretical constitutional scholarship than I have in the last five years, which gives him an advantage in this discussion.

Taking 17 first:
Legislators can't be "dishonest" in the same way as judges, because no one, not even the Federalist Papers, presupposes that a legislator is neutrally adjudicating questions. If your question is meant to ask whether legislators are voting for or against legislation for reasons other than their stated ones, I'll concede that some people who voted for caps did it because they believed caps work for social benefit, some did it because they wanted to help doctors and/or insurance companies at the expense of lawyers, some did it because they thought their constituents wanted the result they voted for, and some did it as a favor for friends or out of spite for enemies; the same set of options applies for those legislators who voted against it. (As a side note, this is a good argument against trying to divine "legislative intent.") But when a legislator puts herself in front of the voters, the voters are free to choose whatever criteria they wish to evaluate the legislator, so "honesty" is amorphous: some voters will vote based on society-wide interest, others will vote on a narrow self- or group-interest, still others will vote based on ethnic or party solidarity or because they don't like the uppity nature of a candidate's spouse. So what does it mean to be "dishonest" in that scenario?

If we believe in the rule of law, the judicial officer should be evaluated by their adherence to the rule of law, so the "dishonesty" comparison is, at a minimum, not apples-to-apples. We don't want judges to be evaluated by random criteria. It comes down to: Can and will this judge neutrally apply principles of law to the cases that come before her? Because of this ideal, there are social norms that prohibit judges from being openly results-based. (Speaking of teeth, judges should be impeached when they clearly violate the norms: e.g., Justice Douglas and tax cases.) The clearer the rules, the harder for a judge to hide results-based decision-making behind a facade of neutrality, deterring dishonesty at the margin.

And it's the marginal judge that makes the argument: the fact that some judges will be incorrigibly dishonest doesn't mean that it's not desireable to make it more difficult for judges to be sneakily dishonest; Ferdon epitomizes the argument, but it's hardly the sole basis for it. Some scofflaws will never be deterred by incarceration, but that's no reason to say incarceration has no deterrent value.

On to 16:
I'm not as familiar as I should be with the work Adrian Vermeule has done since he vacated the twelfth-floor law firm office next to mine, so I withhold comment on whether I'd go as far as you suggest he does.

If legislators pass a Sherman Act that hands complete discretion over to the courts, judges have to do the best they can under the circumstances. I'm not suggesting that judges abstain. (However, I will endorse Frank H. Easterbrook, The Limits of Antitrust, 63 Texas L. Rev. 1 (1984), on the question of how judges should approach the Sherman Act.) But I am suggesting that society not give unfettered discretion to the courts in the first place. For example, the but-for world is difficult to determine, but the courts quite arguably did more harm than good for the first eighty years of antitrust enforcement. Bork, The Antitrust Paradox. But that's a side-debate.

Vis-a-vis the Fourth Amendment, (1) I don't think "probable cause" is that complex of a legal concept; (2) bright-line rules as to "reasonableness" are preferable to balancing tests, especially given the occasional split-second decisions that must be made; and (3) even here, I think there's a distinction between that which is wise and that which violates the Fourth Amendment. E.g., Hedgepeth v. WMATA (D.C. Cir. 2004) (Roberts, J.).

Post a comment

If you have a TypeKey or TypePad account, please Sign In

Recent Comments
Popular Threads
Search The Glom
The Glom on Twitter
Archives by Topic
Archives by Date
January 2019
Sun Mon Tue Wed Thu Fri Sat
    1 2 3 4 5
6 7 8 9 10 11 12
13 14 15 16 17 18 19
20 21 22 23 24 25 26
27 28 29 30 31    
Miscellaneous Links