03.12a. Harvard's Lack of Used Bookstores; Norms; Procedural Protections; Michael Berube and Erin O'Connor on the Nona Gerard Case at Penn State. I'm at Harvard today for a conference for the Handbook of Law and Economics. Richard McAdams and I are writing the chapter on "Norms in Law and Economics." I note with sadness that what I think was the only true used book store in Harvard Square, the one down on the triangle block at Bow Street, has closed down since I last visited here (the Harvard Bookstore has used books in the basement, but it's basically a new book bookstore). Is this a nationwide trend? There's only one used book store in Bloomington now of the classical style, Caveat Emptor on the Courthouse Square, compared to the three that existed in 1992 when I moved to Bloomington. Last I knew, Berkeley and Chicago still had multiple stores, though.

The various adventures in academic administration that I've been following are helping me get a grip on the importance of norms. What protects academic freedom? My law school friends tell me that they think tenure protection actually is quite weak, as a matter of contract law. They like procedural protections-- a requirement that the university have lots of hearings and committee votes before it fires a tenured professor. This is how we do things in criminal proceedings, too. The 90-page Student Code that I criticized on March 3 is another example.

I don't much like procedural protections, though. For one thing, if the university does have a good reason to strip someone of tenure, and everybody knows it, including the professor, it still has to go through all the procedure. This defect has general application: it is costly to require a procedure if the procedure is unnecessary in 99% of cases. It would be better to have a control method that kicks in only in the unusual cases.

Also, even though they hinder legitimate punishments, procedural protections are not all that much protection for the truly abusive cases. Just think of the Martha Stewart case that I discussed March 11. She had all the protections of criminal procedure-- no small amount of protection-- but that fundamental law was rotten. If you disagree, just think of a hypothetical. Suppose we had a law making it criminal to be Christian. Even with all our procedural protections, the government could go after and convict people like me, because we would be truly and, I hope, obviously guilty. (We could of course argue that the law is unconstitutional, but that is a substantive argument, not a procedural one.)

Or look at the Nona Gerard case at Penn State. She has complained in the newspaper that Penn State did not follow the correct procedures, but I bet she'll lose with that argument. Penn State did follow essentially the correct procedures in firing her, and any quibbles she may raise about the length of the hearings or whatever are about things that wouldn't have affected the result. Clearly, the Administration was out to fire her, and they were willing to incur the time and effort costs of the procedural protections. Those costs would deter some firings, but not this one.

This is worth explaining more generally, I think. The essential effect of procedural protections is to raise the cost of firing someone. If we have procedural protections with no substantive protections, then anybody can be fired; it is just a matter of paying the cost. We might for example, have the substantive rule be "A professor can be fired if she does something the President dislikes," and the procedural rule be "To fire a professor requires that the President attend at least 10 2-hour meetings of 10 different committees of faculty, who will vote on whether they believe the President dislikes the professor in question." If faculty committee members are honest, and the President says he dislikes the targeted professor, the committee would have to vote to fire too, according to the substantive rule, just as the Martha Stewart jury, being honest, had to vote that she indeed lied to the government.

This means that the procedural protection boils down to cost. We could summarize that cost in dollars. Suppose the 10 2-hour meetings costs the University $20,000 in faculty and administrator time. The result, then is similar to if we had as a procedural protection that the University may fire any professor if the President is willing to authorize the burning of 200 hundred-dollar bills. (I will pass over the important difference that the President may value his own time loss more than the hundred-dollar bills or the faculty committee member time loss.)

I've run out of time, though, so the end of the discussion of procedure vs. substance, Professor Berube, and Erin O'Connor will have to wait.

[in full at 04.03.12a.htm .      Erasmusen@yahoo.com. ]

To return to Eric Rasmusen's weblog, click http://php.indiana.edu/~erasmuse/w/0.rasmusen.htm.