General Motors Corp.'s board gave a cool reception to the idea of acquiring Chrysler LLC after GM's top management discussed the matter at a meeting last week, people familiar with the matter said.

                                - Wall Street Journal, Monday, October 13, 2008, p. B1

I have no idea if GM's acquiring Chrysler is a good idea or a bad idea.  A so-called "cross-town" merger would certainly be unprecedented, at least if you count the last eighty years or so (well, Chrysler acquired American Motors but that barely counts - it got the Jeep, but it had to take the Pacer and the Gremlin along with it).  What struck me about this was the generally bad press boards get.  Here's one apparently doing exactly what it should do (and, I think, most do), which is apply judgment.

My broader question has to do with the regulation of the judgment-making process, particularly when you consider reactions like this from thoughtful people like my friend David Hoffman over at Concurring Opinions:

But what can we do about "mere" incompetence that imposes severe social costs? As I've been telling my corporate law class these last few weeks, current doctrine provides very weak to nonexistent remedies for negligence, no matter how widely its effects are felt. This doctrine is based on an empirical intuition about the relationship between law, risk-taking and entrepreneurship. That is, the law assumes that business risk-taking requires a special kind of legal immunity, and is particular liable to be badly judged in hindsight. I suspect that both of these assumptions are wrong. . . .

I left Dave a comment over there to the effect "would that the lawyer's notion of cause-and-effect ruled the world!" I'll expand on what I meant below the fold.

What causes something bad to happen is a tricky question in science, philosophy, and the law.  H.L.A. Hart and Anthony Honore wrote a whole book about it, and Michael Moore has one coming out later this year. Here, in something of a nutshell, is my problem with the idea of close after-the-fact scrutiny of judgment.  The philosophical revolution of the Enlightenment that underlies the rise of science (physical and social) is the general rejection of metaphysics as the source of why one thing causes another.  Hume's contribution, as Hart and Honore observe, was to say there is no mystical or divine hand "causing," for example, a billiard ball to go off in a particular direction when struck in a particular way.  What we call cause-and-effect is really just an expectation based on our observation of regularities in the world.  That's really the basis for all science:  we observe and attempt simply to subsume our observation under rules that are as general and universal as possible.  (The project of Hart and Honore is to say, well, there are infinite "but for" causes - what pops out of the mix sufficiently for us to label one particular relationship as "the" cause.)

As I've written, law, it seems to me, is but another model by which we subsume the regularities of experience under rules, and to do it without metaphysics.  (See Fred Schauer on rules as entrenched generalizations.) In everyday life, we rely without much thought upon what somebody tells us that a third person said, but the law looks at that and says: "there's too much danger of a miscommunication along the way to allow that statement to be evidence in an adjudication."  In other words, y=f(x), where x is the hearsay statement and y is truth, and there is something significant less than a 1:1 relationship.  In the law's model, we reject x unless another rule (an exception to the hearsay rule) dictates otherwise.

Apart from the observed phenomenon of hindsight bias, the problem with seeing a bad result and subsuming it under a rule is over-determination.  Any number of patterns "cause-and-effect" might have led to this result - why are we focused on just this one?  So if the GM board approves a cross-town merger, and it fails, do we assign blame based on the rule "a merger of two bad businesses simply creates one larger bad business"?   And for the purposes of law, anybody who allowed that to happen must have been incompetent?  On the other hand, the problem facing the GM board today is under-determination - there are myriad "cause-and-effect" patterns that could explain how we got here.  How do we choose a cure that addresses the correct cause? 

I can explain this a different way.  There is a commonly understood rule that at trial when cross-examining a witness, you do not ask a question to which you do not know the answer.  If you follow that rule, you will likely never make a mistake, but as great cross-examiners know, you may miss out on some gold as well.  Great cross-examiners have an intuitive sense about when they can ask a question to which they don't know the answer, but it's a riskier business.  Let's assume you are indeed a great cross-examiner.  You sense there is an very strong chance that by asking the question to which you don't the answer you win, but there's also the chance you lose.  It's the middle of an examination, so you don't really have time to consult your client. It's now or never.  If your concern is avoiding malpractice, follow the general rule because in hindsight, somebody is going to disagree with your assessment of the odds.  If your concern is winning cases, go ahead with the question.

Once again, Thomas Friedman has an apropos commentary.  In his October 14 column, he laments the distance that got interposed between a mortgage borrower who ought to have shown real income and a good credit record into a world in (my words slightly tweaking and interpreting his) in which somebody with a financial algorithm bought up a bunch of loans, made a calculation of the risk (i.e. subsumed the regularities of the world in a probablistic model), applied a discount rate, and sold the instrument to an Icelandic banker.  Friedman quotes Dov Seidman, the CEO of LRN (and a Harvard Law grad, by the way), to the effect that "engineering money from money" overcame good judgment.  Friedman therefore concludes:  "We need to get back to collaborating the old-fashioned way.  That is, people making decisions based on business judgment, experience, prudence, clarity of communications and thinking about how - not just how much."

It seems to me that the inherent wisdom of the business judgment rule is that a decision requiring business judgment, experience, prudence, and clarity of communication is too under-determined to be placed within the system of perceived regularities (i.e. cause-and-effect) that is law's model of the world.  The problem, in my view, is that we've given over too much judgment to the algorithmists already; the last thing we want to do is address the problem by encouraging even more formulaic decision-making at the risk of hindsight liability.

Corporate Governance | Bookmark

TrackBack URL for this entry:
https://www.typepad.com/services/trackback/6a00d8345157d569e2010535819899970b

Links to weblogs that reference How Do You Regulate Good Judgment?: