Reinhard Tietz

Ultimatum bargaining behavior: a survey and comparison of experimental results

In an ulttmatum bargaming game players 1 and 2 can drstribute a positive amount of money m the following way: first, player 1 determines his demand which player 2 can then etther accept or induce confhct, i.e. player 2 faces the ultimatum either to accept player l’s proposal or to have no agreement at all. Experimentally observed ultimatum bargaimng decisrons with amounts ranging from 0.50 to 100 German marks are statistically analysed. The demands of player 1 are compared with the acceptance behavior of player 2 with the help of consistency tests in which a subject has to decide in the positron of both players. Finally, we constder ulttmatum bargaining games with more than just one round where, except for the final round, nonacceptance does not cause conflict but another round of ultrmatum bargainmg for a smaller cake.

Bounded Rational Behavior in Experimental Games and Markets

I. Experimental Economics.- Experimental Economics: Ways to Model Bounded Rational Bargaining Behavior - Introductory Remarks to the Fourth Conference on Experimental Economics.- Experimental Economics: A Psychological Perspective.- II. Unilateral Decisions.- Normative and Individual Strategies in Social Dilemmata.- Learning to Make Good Predictions in Time Series.- A Preliminary Analysis of a Large-Scale Experimental Investigation into Consumption under Uncertainty.- Mental Representation in Multistage Decision Making.- III. Bilateral Decisions.- Stability and Outcome Tradeoffs in Asymmetric Dilemmas: Conditions Promoting the Discovery of Alternating Solutions.- The Influence of Social Orientation and Generalized Expectancies on Decision Making in Iterated Experimental Games.- Ultimatum Bargaining for a Shrinking Cake - An Experimental Analysis.- Bounded Rational Strategies in Sequential Bargaining: An Experiment and a Learning by Evolution Strategy.- Semi-Normative Properties of Bounded Rational Bargaining Theories.- Modeling Mediator Behavior in Experimental Games.- IV. Auctions and Markets.- Dual Ceteris Paribus Comparisons: Some Initial Thoughts on Laboratory Isolation of Equilibrium Forces in Auction Markets.- Learning in Common Value Auctions.- Inflation and Expectations in Experimental Markets.- Modeling Investment Behavior in an Experimental Market.- V. Coalition Formation.- Order of Strength and Exhaustivity as Additional Hypotheses in Theories for 3-Person Characteristic Function Games.- The Effects of Justice Norms in a Bargaining Situation.- Modeling Coalition Formation in Inessential Probabilistic Games.- Sequential Games of Status: A Replication.- Equal Share Analysis for Location Games.- Limits of Rational Behavior in Cooperatively Played Normal Form Games.- Revealed Aspirations and Reciprocal Loyalty in Apex Games.- VI. Bibliography.- Selected Bibliography of the Participants Contributions to Experimental Economics.- Author Index.

Ultimatum Bargaining for a Shrinking Cake — An Experimental Analysis —

In an ultimatum bargaining game two players have to divide a given positive amount c of money. First player 1 demands how much of the ‘cake’ c he desires for himself. Then player 2 can either accept or reject this proposal which shows that player 2 faces an ultimatum. If 2 accepts player l’s proposal, player 1 gets what he demanded and player 2 the residual amount. In case of rejection by player 2 we follow the experimental procedure of Binmore, Snaked, and Sutton, i.e., there is another round of ultimatum bargaining with exchanged roles about a “cake” c’ with c’ < c. If also the second round of ultimatum bargaining ends with a rejection, the game ends with 0-payoffs. Our study differs mainly in two aspects from the original experiment of Binmore et al. The monetary payoffs are higher and the game theoretic solution is more extreme. To explore experience each subject was engaged in two subsequent games. One main result is that contrary to Binmore et al. the game theoretic solution has nearly no predictive power. Whereas in one round-games the major consideration is to protect an unfair agreement by sufficiently high cost for choosing conflict, limited rationality will require to analyse the more complex two round-games in a completely different way, e.g. by considering the amounts c and c’ as resources of that player who has the right to propose.

Semi-normative theories based on bounded rationality

Abstract An attempt is made to classify descriptive economic theories by various qualitative properties. To close the gap between normative and descriptive theories one can not only postulate a good descriptive quality of a normative theory but also sufficient normative qualities of a descriptive theory. In particular, descriptive theories which fulfill the assumptions of bounded rationality must have certain semi-normative properties. The degree of rationality of the behavior described depends an such properties as the consistency of expectations or the type of effect of the payoffor goal-variable has on strategic variables. For example, the payoff function should have a concave relationship to the aggressiveness of behavior in order to favor equilibria. The areas which favor equilibria are able to explain why experienced behavior is moderate. They are also important for stabilizing behavior in ‘rule equilibria’. Behavior within rule equilibria is observed more frequently than behavior that lies outside them. From these general postulates it follows that the building blocks of a theory of bounded rationality should be based less on one-shot experiments than on repeated experiments where mature behavior can be observed. Inducing cognitive dissonances in the planning phase of the decision process improves the consistency of expectations and the rationality of decisions. In order to detect those parts of behavioral rules which favor equilibria, it is sometimes more important to observe and model infrequent behavior of a few subjects in extreme situations than to explain average behavior of many subjects in typical situations.

Adaptation of Aspiration Levels - Theory and Experiment -

Stimulated by the work of Herbert A. Simon, Reinhard Selten has concerned himself with aspiration levels and bounded rationality since the beginning of his academic career.2 His joint paper with Heinz Sauermann “Anspruchsanpassungs-theorie der Unternehmung” was published already 1962.3

Profit Sharing in an Asymmetric Bargaining Game

It is a stylized fact of many experiments on the ultimatum game and alternating offer bargaining games that play deviates from the game theoretic solution toward equitable allocations. Evidence for this result are the surveys by Guth and Tietz (1990) and Thaler (1988) as well as the more recent ones by Roth (1995) and Camerer and Thaler (1995).

Experimental Economics: Ways to Model Bounded Rational Bargaining Behavior

Experimental Economics is an interdisciplinary research method in which mainly economists and psychologists cooperate in analyzing human decision behavior. Testing hypotheses and constructing theories are the two main tasks of this discipline.

SEMI-NORMATIVE PROPERTIES OF BOUNDED RATIONAL BARGAINING THEORIES

It is postulated that descriptive bargaining theories should have semi-normative properties. They concern the influence of tactical variables, the first demand or the conflict threat; on the agreement. A computer simulation shows that the dynamic aspiration balance theory has some semi-normative properties such as the convexity of the payoff profile over the aggressivity of the first demand. This profile is shifted and modified by aggressive threats. Properties of other bargaining theories are discussed.

On the Structure, Stabilization and Accuracy of the Decision Process

For an improvement of human relations within and between nations a better understanding of interactive human decision making is needed. In economic research concerned with decision making there are two contrasting approaches: the classical theory of the absolute rational decision maker, who maximizes his utility function, and the behavior oriented theory of the limited rational decision maker, who searches for a satisfying alternative.

Expectations and Rational Actions in an Experimental Financial Market

Models of economic behavior under conditions of uncertainty typically describe the influence of subjective beliefs about future outcomes on market behavior. Tests of such models are limited by the lack of hard information regarding expectations. We report here on an experiment in which we obtain a direct measure of expectations regarding a particular natural event and detailed information regarding market transactions in a futures contract using that event as a fundamental. This unique coincidence of direct observations on expectations and concurrent market behavior permits an assessment of the consistency of behavior and expectations which is not possible from any other data as far as we know.