ACCORDING TO AN EMAIL from Bloomberg.com reviewer Jim Pressley "Your book has incredible relevance on a day when U.S. stocks have fallen to their lowest levels since 2003." He was talking especially about the Ten Top Tips that I offer for using game theory strategies to promote cooperation.
MANY ECONOMISTS have commented on the need for rebuilding trust (in this case, between financial institutions). The trust needs to have a genuine basis of credible commitment from all sides, which means (as the economists have pointed out) that in this case that there needs to be transparency in revealing the true value of the assets used to back loans. But how might it be achieved?
GAME THEORY has a lot to say on this topic, and many of my Ten Top Tips could come into play. Here is one suggestion, based on my Number Ten Top Tip - divide large groups into smaller ones (and build co-operation up from this base). Game theory shows that very large groups of individuals or institutions have a lower chance than smaller groups of coordinating their actions to produce a cooperative outcome, because repeated interactions between individual pairs within the group are less common, and other restraints on cheating are often less powerful. One way to re-establish trust, then, is to begin with groups consisting of a small number of institutions. Trust is easier to establish in such a small group. Maybe, for example, a pair of institutions will be more willing to reveal their assets to each other because they have a relationship that extends beyond the market. If one of them had a similar relationship with a third institution, then the group could be enlarged from two to three (cf. my Tip Two - bring an extra player in) and so on progressively. If, at any stage, trust in one of those institutions turned out to be misplaced, my Tip One - Stay if you Win, Shift if you Lose comes into play, which in this case means cutting the offending institution out of the network. Keeping the network to a finite size would also mean that repeated interactions would be more common (Tip Three - set up some form of reciprocity), and an institution’s reputation within the network would suffer if it did not deliver (Tip Four - restrict your own future options so that you will lose out if you defect on cooperation). Finally, negotiators in the not-too-distant future may be able to use the remarkable properties of quantum computers (as I describe in chapter eight of "Rock, Paper, Scissors) to “read each others' minds” – the ultimate in transparency, and perhaps the ultimate in removing incentives to cheat.
ALL OF THIS is not to say that game theory provides complete answers, but it does suggest sensible directions AND PROVIDES GOOD REASONS FOR FOLLOWING THEM, BASED ON THE REAL WAY IN WHICH PEOPLE BEHAVE IN PURSUIT OF THEIR OWN BEST INTERESTS. Maybe this is the best that we can hope for.
FOR MORE BACKGROUND, see my November 9 post:
0 comments:
Post a Comment