Streetwise Professor

Matt Ridley, discussing a paper Slavisa Tasic presented at the  Instituto Bruno Leoni conference at which Renee also presented*, reminds us that regulators can have cognitive biases too.

This is a pretty unexceptional point.  But it raises the question.  If both individual decision makers in markets and regulators are both subject to cognitive biases, isn’t it kind of a push?

I say no.  Decisively no.  For two reasons.

First, the feedback mechanisms are different.  Individual decision makers are subject to market feedback, and this feedback is often quick and ruthless.  For instance, an individual trader who uses a biased decision rule is likely to lose money, and often a lot of money.  Indeed, those who get rich in these markets often do so at the expense of the fools–or should I say, the cognitively biased/impaired?  Evolutionary rationality often trumps individual irrationality.

In contrast, politicians and regulators face very weak feedback, and often no feedback or perverse feedback.  To take an extreme example, what feedback will Barney Frank or Chris Dodd receive if something buried in Frank-n-Dodd runs amok and causes substantial damage in the market?  Answer: nada.  Similarly, a regulator that makes a boneheaded decision is largely immune from being fired, and may well be working for Goldman by the time the bad consequences of his or her decision comes to light.  Evolutionary mechanisms don’t work with anything near the same force to weed out the irrational and the stupid in political and regulatory contexts.

Second, regulatory and legislative mistakes are often systemic in their effects, whereas individual decision errors are less likely to be so.  A regulatory or legislative mistake affects everybody subject to the regulation or law.  For instance, if it turns out that forcing greater use of clearing is actually a bad idea, but regulators and legislators acting under the delusion of competence mandate it anyways, the entire market bears the brunt of that mistake.  And the cost of that can be quite extreme.

In contrast, although herding behavior (correlated cognitive biases) can occur, and may contribute to bubbles and runs and other undesirable outcomes, a lot of individual biases tend to cancel out.  (I would also argue that many herding phenomena that are attributed to biases may in fact be rational responses to perverse incentives.  For instance, the fact that a lot of financial institutions loaded up on AAA mortgage CDOs and PIGS debt was likely the result of the perverse incentives of the Basel system than some mass delusion.)

In other words, laws and regulations tend to put all the eggs in one basket.  Markets and other emergent orders tend to be more diverse, and diversified, and less vulnerable to the errors or biases of an individual or a small group of individuals.

So the Tasic paper that Ridley writes about is a useful antidote to knee-jerk behavioralism that is often used to rationalize the need for more regulation.  But it is only a good start.  Once you understand that both market agents and regulators/legislators are cognitively imperfect, you need to go farther and examine what the consequences of those imperfections are.  To do so, you need to focus on differences in feedback and diversity across regulatory/political mechanisms on the one hand and market mechanisms on the other.

* This was Renee’s first conference presentation.  Papers were selected competitively.  Hers was on the parallels between the forces leading to commodity regulation in the 1930s and today.  Not only was she the only presenter at the conference without a PhD, she doesn’t have any degree yet, being in the last year of a 5 year joint MA/BA program.  Dad is very proud.