Prolapse Can Spotting Of Blood

Prediction and factum in Microeconomics I: rationality and Operating

In this sequence of posts I will (try to) make a review of some of the paradoxes 'famous' of microeconomic theory. That is, cases in which the predictions arising from it are inconsistent with those observed empirically. In this post, it's the turn of the ultimatum game.

Prelude

At the risk of oversimplify important issues of epistemological status, can be certain that the economy-as-discipline studies human behavior in a broad sense: much of the research in this area seek to define, identify and verify (empirically or experimentally) the logical implications derived from certain basic tenets or axioms used as a starting point. This exercise serves, among other things, to verify the validity of the axioms or assumptions (that is, how relevant to determine or explain the behavior of the phenomenon under study).

Undoubtedly, the axioms of rationality of the agents belong to the hard core of decision theory modern, becoming the ultimate reference frame (either by difficulties in modeling 'irrationality 'or the convenience of thinking in terms of deviations from a common denominator). However, reporting requirements and capacity calculation involving the "rational" are "in the best-a stylized characterization of the decision-making framework under which economic agents operate, so there are several arguments for doubting the importance or relevance of the course.

In this sense, models of bargaining or bilateral negotiations are an excellent theoretical scenario to test the rationality of agents, because through a simple game can compare the predictions of theory the actual interaction of those who lend themselves to this type of experiments (as a footnote, note that curiously tend to be students of economics).

(To motivate a little post, let me note that several real-life situations can be represented as a bargaining model, for example the interaction of actors in decentralized markets, the housing market The market for lemons the work, etc.-or explicit resolution / institutional conflict, with specific applications in labor economics and what is known as political economy .)

Prediction

The most natural example to illustrate the late-exposure is the game the Ultimatum , the sequence describe below

• Two players (1 and 2) must be allocated a sum of money D;
• First, player 1 must make an offer (s , 1 - s ) to player 2, where s is the fraction of the sum D which is player 1;
• Once the offer, player 2 can accept (in which case the money is distributed as agreed) or reject (in which case both players are left without any money).

The classical result of game (Rubinstein, 1982 ) shows that the equilibrium strategy is more robust that player 1 is left with all the money (s = 1 - 'epsilon') and Player 2 accepts. And at this point, there are two key assumptions are: first, all players looking to maximize their own pay (baptize this "homo-economicus "), also have rational preferences and know that the other players known to follow this type of behavior. latter is called common knowledge of rationality .

balance Intuition proposed is simple: if player 2 rejects the offer runs out of money, therefore a "homo-economicus " will not reject offers that will leave a positive sum, however small it may be. Note that in this context can not be made credible threats . So player 1 can comfortably provide s = 1 - 'epsilon' .

Factum

Without

But this is not the way the negotiations are resolved in reality. Güth, Schmittberger, Schwarze (1983 ) conducted the first experimental study of this simple game, and found that the average supply of players 1 was about 37%! of the total, which is clearly in short with our prediction of about 100%. Moreover, almost half of the "Players 2" rejected offers of less than 30%.

Naturally, this paradox drew the attention of the researchers, who tested several assumptions in order to reconcile the prediction the factum. A first online attacks the paradigm of rationality, adding restrictions to the computing power of the agents (bounded rationality). Even in a simple game like this, it may be possible that there is "gotcha" on the bid predicted. In this vein, one could argue that trying different repetitions of the same game, players 1 end up "learning" that can enhance their offerings to reach equilibrium, although this is not a recurring pattern in experimental studies.

Another solution to this paradox defends the traditional concept of rationality, but relaxes the notion of " homo-economicus "incorporating preferences for equality / inequality in the distribution. That is, the "objective function" of the agents is not only its own output, but also incorporates the results of the other players. So when a player 2 rejects an offer as insufficient is "punishing" the player 1 for his selfishness. In this vein, Levine (1997 ) inferred distributions of altruism and selfishness from the experimental results of this type of game. Surprisingly, these distributions fit quite well the results of other games (for example, funding a public good).