Friday, October 30, 2015

Biases and capture after the book by Carpenter and Moss

We can speak of a number of deviations from bounded rationality that affect the behavior of regulators and agents who interact in the political arena:
-The difficulties of optimizing behavior, either because regulators stop at a satisfacing performance without reaching the optimum, or because they simply behave adaptively, for example acting only when there is an event of great visibility. As the referees in sport, in normal times the regulators would not be carried away because of the "omission bias."
-The existence of intrinsic preferences, either because regulators are moved by an ethos of public service, or a concern for their reputation, or to remain faithful to a rooted belief that comes from the educational process.
-Expert biases, such as confirmation bias, or overconfidence in their own expertise. For example, in Chile the expert engineers who designed the system called Transantiago for the buses in the capital, did not ask the opinion of politicians or ordinary citizens, and this probably led them to overlook the enormous transaction costs that the sudden introduction of a whole new system implied, resulting in a huge social and political scandal.
-The importance of the processes, so that those involved in a particular reform may be more inclined to accept its results, regardless of the exact outcome, if it has been preceded by a fair and participatory process.
-The possibility of cultural capture, because of identity reasons, or again regulators' beliefs forged in certain training institutions, or the status in relation to a particular group or market or professional relationships in the context of a particular social network.
Some biases bring regulators closer to the possibility of regulatory capture, and others away from it. Something similar could be said regarding the possibility of achieving credible commitments. Non measurable cultural aspects related to institutional quality and prevention of capture are closely related to those sources of bias, and take on greater importance than in more traditional analyses.
It is interesting that the possible mechanisms for preventing capture and minimizing its effects are very similar to de-biasing mechanisms (see the excellent book on capture by Carpenter and Moss): have mechanisms that make it necessary to find evidence that "dis-confirms" the held hypotheses, have internal mechanisms of devil's advocate, systematic assessment of ex post decisions, transparency and the obligation to explain decisions, etc. Institutional diversity itself can be a good recipe to experiment and innovate before generalizing possible solutions to the problems of regulation and antitrust in a context of uncertainty and technological complexity.

No comments:

Post a Comment