Thursday, February 10, 2011

Prisoner's Dilemma

Long time ago some very smart thinker created a concept called the Prisoner's Dilemma. The model was there to show a plausible way to analyze the competitive interaction of two people and set who were individualistically looking for payoffs, the two agents being in a situation that somehow hinted for ruining the other agent's possibility to any gains as a result of the personal-egotistic nature of each one agent. Moreover, the Prisoner's Dilemma turns into a moral about how seeking for the personal benefit will result into the ruin of everyone in a real life situation, given, of course, that that situation's conditions match those of the model.

The above said, it must be noted that the Prisoner's Dilemma is regarded as a masterpiece of counter-intuitive analysis, and considered a concept peer to that of minimax analysis. I applied my own version of the acid test to the Prisoner's Dilemma, and found out that a lot of its power comes out of the minds of those who are beiong introduced to it rather than being an essential part of the formulation of the model itself. To say this in simpler terms: the model is incomplete because it relies on the emotions of the analyst rather than on its own structure. Allow me to elaborate.

Suppose there are two small firms dedicated to the development of new energy sources: Xenergy and Y-Fuel, and that both firms belong to the same market, and have some interactive dynamics that will restrain them from being held by two shareholders, making the matter of the share value a very important one. Furthermore, there is a duopoly in the market for energy generation, so that AlphaCo and BetaPower are the only two major firms that might be interested in acquiring the new small firms. Now, we mentioned that interaction will affect the performance of the small firms. In detail, the conditions, summarized in the table below, are:


Beta X Beta Y
Alpha X 2, 2 0, 3
Beta Y 3, 0 1, 1


1) If AlphaCo and BetaPower both of them decide to purchase share packages of Xenergy, then profits coming out from the total annual activity will be of 4 billion dollars, allowing each one major firm to receive 2 billion dollars of net profits. This seems to be appealing to both firms, for the amount earned is huge.

2) If AlphaCo decides to buy Xenergy, and BetaPower buys Y-Fuel, then the net profits in the market will be of only 3 billion dollars, and Xenergy won't get any of them, therefore leaving AlphaCo with zero profits at the end of the year, and BetaPower with round three billion dollars. While this scenario is really appealing to BetaPower, it threatens to harm AlphaCo level of profits.

3) If AlphaCo buys Y-Fuel and BetaPower buys Xenergy, it will be AlphaCo the one to make the three billion dollars, and BetaPower to get zero profits. An scenario appealing to AlphaCo, but dreadful to BetaPower.

4) Now, if both of them decide to buy shares from Y-Fuel, then the performance of the energy market will be that of 2 billion dollars, to be split between AlphaCo (one billion dollars) and BetaPower (one billion dollars).
And then which one should we expect to be the definitive scenario for this interaction? Well, according to the conventional approach for the prisoner's dilemma, we should expect both firms to try to outsmart the other one and generate a (suboptimal) outcome such as Y-Y (AlphaCo will buy Y-Fuel under the assumption that BetaPower will likely buy Xenergy, and BetaPower will buy Y-Fuel under the assumption that AlphaCo will buy Xenergy). Each one firm receives the suboptimal profit level of 1 billion dollars at the end of the year. Now, let's review this conflictive interaction (namely, a "game"), with two different, simpler individuals. Let's say Al and Bob, two young brothers.

So one day Al and Bob's uncle Ernie comes with two two-sided cards, each card having a white side and a black side. Uncle Ernie gives a card to each one and proposes a game, that if each one chooses the white card, he'll give two dollars to each one, if one chooses the white card and the other the black card, the one with the black card gets 3 dollars and the one with the white card nothing. Finally, if each one gets the black card, each one gets one dollar from him. The kids are taken to two different rooms, so that they won't be able to talk to the other. Al's thinking process goes as follows: if I choose the white card, then I am giving two choices to Bob. Choice 1: that he picks also a white card, so that both of us receive 2 dollars. But he will also have a second choice, to pick a black card, in which case I do not receive anything and he gets three dollars. And if I pick a black card and he picks a white card, then he doesn't get anything, and I get three dollars. But if he decides also to pick the black card, then I am getting just one dollar.

No comments:

Post a Comment