Information Acquisition Games in Cournot MArkets: A Three Stage Approach
by
Eliane Catilina
University of Virginia

Information Acquisition Games in Cournot Oligopolistic Markets: A Three Stage Approach, by Eliane Catilina.

Extended Abstract Information Acquisition in Cournot Oligopolistic markets has mainly being modelled as a two-stage game in which in the first stage firms decide whether or not to purchase information and in the second-stage choose output. It has also been shown that firms have no incentive for information sharing (first-stage) when output is competitively determined (second-stage). In this paper I argue that the current research on information acquisition fails to identify an intermediate stage in which, after deciding to purchase information and observing competitors' decision, firms decide whether they want to collude and share the cost of information acquisition. This paper presents a three-stage game approach to information acquisition that captures this "cooperation" stage. The game runs as follows:

(i) First-stage: Upon observing the information acquisition cost, each firm decides whether or not to purchase information. This decision is not binding, nevertheless, revealed to the competitors.

(ii) Second-stage: Firms that, in the first-stage, declared that would become informed have now to decide if they want to form a collusion to share the cost of information acquisition. The information is purchased either by individual firms or by the collusion.

(iii) Third-stage: The collusion is dissolved and (informed and uninformed) firms simultaneous and individually set output.

The main findings of the paper are as follows: 1. If the information acquisition cost is sufficiently low, firms do not have incentive to share the cost of information acquisition and the outcome of the three-stage game is similar to the two-stage game.

2. If the information acquisition cost is sufficiently high, all firms remain uninformed (first-stage) and so a collusion cannot be formed in the second-stage. Again, the outcome to the three-stage game is similar to the two-stage game.

3. The solutions to the two and three-stage games are substantially different at the cost range in which, in the outcome of a two-stage game, firms would be asymmetrically informed. In the three-stage game, within the above mentioned cost range, we have that:

(a) If the information acquisition cost is sufficiently low firms have no incentive to share the cost of information. The information asymmetry persists as in the two-stage approach.

(b) If the cost of information acquisition is sufficiently high, all informed firms collude in the second stage and share the cost of information acquisition. The information asymmetry persists but now the payoffs of the informed firms are higher that what they would be in the two-stage approach.

(c) At an intermediate cost level, in the equilibrium of the three-stage game, all firms decide to become informed (first-stage) and share the cost of information acquisition (second-stage). The information asymmetry is eliminated and there is an "involuntary" perfect information sharing. The payoffs of the firms that would otherwise be uninformed (two-stage approach) are now higher. On the other hand, the payoffs of the firms that would be asymmetrically informed (two-stage approach) is now lower as there is a perfect information sharing.

4. In the outcome of the three-stage game firms acquire more information than in the two-stage game as the "all-firms-informed" outcome arises as an equilibrium in more than one cost range while in the two-stage approach this outcome arises as an equilibrium only if the information acquisition cost goes to zero.

Date received: July 18, 2000

Copyright © 2000 by the author(s). The author(s) of this document and the organizers of the conference have granted their consent to include this abstract in Atlas Conferences Inc. Document # cafl-40.