Sunday, February 17, 2008

The Economics of Cooperation

This post is going to require at least a basic knowledge of the prisoner’s dilemma (PD) and moreover what is so special about the iterated prisoner’s dilemma. However, I hope it has some interesting points even for those who know nothing about it, and cannot be bothered to do any research.

The IPD allows punishment and reward for behaviour, and it allows participants to learn and create a reputation. These factors were enforced when Robert Axelrod[i] conducted experiments where computer programs were pitched against each other in an IPD. The winner was Anatol Rapoport’s program TIT FOR TAT, which has yet to be beaten by another single program (a multiple program entry involving slaves and masters did beat it), even when tested against hundreds of programs designed specifically in the knowledge of its success. The triumph of TIT FOR TAT was thought to be down to the following characteristics (as stated on page 110 of ref. (1)):
1) Don’t be envious
2) Don’t be the first to defect
3) Reciprocate both cooperation and defection
4) Don’t be too clever
These points outline how successful economic behaviour should be conducted, and relies on the knowledge that the PD is not a zero-sum game, that is whatever someone gains the other does not stand to lose, or vice versa. Therefore, by point one, you should not worry if your immediate competitor is beating you, it’s the long run that matters. By point two you should not be the first to defect – again in the long run the punishments (loss of reputation for example) would outweigh the short term benefit. Point three shows how you should be retaliatory and forgiving, by this you will not be abused by exploitative institutions, and you will quickly forgive single instances of provocation (or in TIT FOR TAT’s case, any instance - another of its virtues was its short memory). Point four shows how you shouldn’t be too clever when evaluating economic behaviour, for instance in a “zero-sum game, such as chess, we can safely use the assumption that the other player will pick the most dangerous move that can be found, and we can act accordingly.” In this situation it pays to be as complicated and multifaceted in an analysis as possible. However, your opponent your economic behaviour effects is not out to defeat you. Also, if you are too complex your opponent may think you are unresponsive and acting randomly.

Indeed examples of the PD are found throughout life, and notably, the ensuing theme of all these is that mutually cooperating is of most benefit to everyone. As a last example, take the ‘live and let live’ system that was endemic in trench warfare in WWI, soldiers would only shoot to wound their enemy, in the hope the action would be returned. In this way a system of reciprocity became a natural part of trench warfare and, if cooperation can evolve out of warfare, maybe everyone should take the time to examine the PD and realise the benefits its implications could have for our world.

[i] Axelrod, Robert – The evolution of co-operation. Penguin books. 1984.

No comments: