Experiments on Durable Good Monopoly

Guth, Ockenfels and Ritzberger (1995) reports on experiments in which a single seller was matched with ten buyers. Each experiment consisted of five or six experimental games. In each game there were different numbers of trading periods(two or three) and discount factors. In one experiment, subjects received prior training in analyzing durable-goods monopoly pricing games. The results for untrained subjects were grossly inconsistent with the theory.

They found that prices failed to conform to comparative statics predictions, and prices tended to be much higher than predicted. However, the levels of prices with trained subjects were closer to theoretical predictions, but prices still failed to satisfy comparative statics predictions

Rapoport, Erev, and Zwick (1995) report on bargaining experiments with time discounting, where they incorporated one-sided incomplete information and an infinite time horizon. The treatment variable was the disount rate, which was the same for both buyer and seller and varied from 33% to 90%. The bargaining process ended when either the buyer accepted the offered price or the highest possible discounted profit became smaller than US$1. They observed in their experiments that

1. price offers tend to decline over time, as predicted by the sequential equilibrium(SE),
2. average initial prices are higher the higher the discount factor, contrary to the SE,and
3. for some discount factors the average initial price was above the static monopoly price, contrary to the SE.

The results can be summarized as follows: first, buyers accept offers too soon. This leads to higher payoffs for the seller but it also leads to higher levels of efficiency than predicted. Second, as the discount factor increases, on one hand, first price offers increase, which is opposite to the theoretical prediction, on the other hand prices for the second periods are higher, which is well in line with the theory. The authors in the end conclude that the equilibrium predictions capture very little of the qualitative features of the data. Subjects seem to follow simple rules of thumb in choosing strategies.

Reynolds (2000) in his experiment on durable goods monopoly compared results of single period monopoly (nondurable environment) to multi-period monopoly which has features of durable goods environment. He found average prices to be below the static monopoly benchmark price in all settings. Also initial prices were higher in multi-period experiments than in single period experiments, in contrast to equilibrium predictions.

Cason and Sharma (2001) investigated the different theoretical predicitons of Coase and Bagnoli et al. in an experiment. (For an overview of the theoretical model click here.) They created a discrete demand environment in which a monopolistic seller faces two buyers. The common discount factors are replaced by a common continuation probability with which in case of rejection another price could be posted. This probability varies between sessions with values of 60%, 75% and 90%. The other treatment variable is the information about the demand. In one treatment it was known that buyers’ valuations in the market are 18 and 54 pesos. In the second treatment only the distribution but not the realization of buyers’ values (buyer 1: 54:95% and 18:5%, buyer 2: 18:95% and 54: 5%) was known. Experimental results are partly in accordance with theoretical predictions. Opening offer prices reject the Pacman strategy. Lower opening prices for treatments with higher continuation probability support strategic considerations rather than the conjecture that fairness considerations might have driven the prices below the ones of the Pacman strategy. Demand withholding by buyers were genereally much higher and lead to longer negotiations than predicted.

Cason and Reynolds (2004) presented a two-player,two-period sequential bargaining game with asymmetric information. Buyers and sellers had the same discount factor, which was presented as a continuation probability to the second period. The treatment variable is the continuation probability, which is varied between sessions from 0%, 30%, 60%, to 90%. Cason and Reynolds found significant differences between the equilibrium predictions of the theory and the experimental data. They proposed two models allowing for bounded rationality, which seemed to capture the interaction in the experiments better than conventional models of homo economicus.

Güth, Kröger and Normann (2004) analyze a durable-goods monopoly selling a single unit of a good to a buyer whose value of the good is private information. Compared to other experiments investigating durable goods monopolies the important difference in this paper is the discount factors of the buyers and the seller which differ and remain as private information. Price is determined by the relative differences in the impatience of the bargaining parties. In their paper Güth, Kröger, and Norman derive the closed-form solution of a two-period game and compare it to the experimentally observed behavior. The results are mostly consistent with the predictions.

References

• Cason, T. and T. Sharma (2001): Durable Goods, Coasian Dynamics, and Uncertainty: Theory and Experiments, Journal of Political Economy 109, 1311-1354.
• Cason, T. and S. Reynolds, Bounded Rationality in Laboratory Bargaining with Asymmetric Information (forthcoming)
• Güth, W., S. Kröger and H.T.Norman (2004) "Durable Goods Monopoly A theoretical and experimental study", Economic Inquiry, July 2004.
• Güth, W., P. Ockenfels and K. Ritzberger (1995): On Durable Goods Monopolies -- An Experimental Study of Intrapersonal Price Competition and Price Discrimination over Time, Journal of Economic Psychology, 16 , 247-274 .
• Rapoport, A. I. Erev and R. Zwick (1995): An Experimental Study of Buyer-Seller Negotiations with One-Sided Incomplete Informationand Time Discounting, in Management Science, 41, 377-394.
• Reynolds, S. (2000): Durable-Goods Monopoly: Laboratory Market and Bargaining Experiments, RAND Journal of Economics 31, 375-394.