conducted this experiment at the University of Amsterdam with students that were recruited through a post on a bulletin board. They conducted five sessions with R = 1 and four with R = ∞ with 12 -14 students in each session. Approximately 50% of the students were majoring in economics but no one had any prior experience with this type of experiment. The results of the experiment revealed irrational bids in both cases. In the case of R = 1 (proportional probabilities), the level of dissipation is at the predicted level, and become more concentrated around the prediction during the last ten rounds. However, the theoretical model fails in predicting the ‘excessive dissipation’ where the level of dissipation is persistently higher but still shows a tendency to move towards the predicted level. In the case of R = ∞ (perfect discrimination), the level of dissipation fluctuated around the predicted dissipation level. Further, the average earnings were significantly closer to the Nash equilibrium under perfect discrimination.
Thus, the authors concluded that the rent-seeking model has predictive power. In both cases, rent dissipation is observed and is in line with the theoretical predictions. The experiment also revealed three types of behavior. Some play like ‘gamesmen’ who understand the strategic nature of the game and play accordingly. Others seem confused and randomize their decisions, while a substantial number of players adapt their behavior to earlier …show more content…
(2000) had players make sequential instead of simultaneous bids. Theoretically, the two-stage game gives the first player the opportunity to make a pre-emptive bid, thus leaving the second player with a payoff of zero. However, according to observations, a second-player advantage was observed. The driving factor, as explained by the authors was the opportunity for second players to punish first players, which was costly to both, even more so for the latter. Further, they exploited the ‘fair’ or efficient behavior of the first players. In addition, emotions appeared to play an extensive role as the second players still intensively punished the first in the last round. Hence with this variation, little support for the basic rent-seeking model was found. I believe this follows the experiment conducted by Miller and Pratt as discussed by Potters et al. and is similarly skewed as they used sequential decisions rather than a simultaneous single decision design. Overall, Potters et al. succeeded by using an appropriate design to investigate the two starkest cases. Their conclusion concerning the predictive power of the rent-seeking model given that one allows that people sometimes makes mistakes seems reasonable. Weimann et al. despite the problematic experimental design also showed evidence of different types of behavior with substantial players using emotion in their case. Thus, Potters et al. successfully explored the