Learning dynamics, genetic algorithms, and corporate takeovers

This paper simulates, via a genetic-learning algorithm, free-riding and coo rdination failure when shareholders are confronted with an unconditional te nder-offer bid between the pre-takeover and post-takeover value of their fi rm. The outcomes produced by the simulations offer strong support for the h ypothesis that coordination to tendering strategies permitting offer succes s is impaired by increasing the number of shareholders and the divisibility of share holdings. Further, the outcomes of the simulations closely confor m to the restrictions imposed by the Nash equilibrium hypothesis. When the number of shareholders and the disability of shareholdings are both small, the aggregate outcomes of the simulations converge to the aggregate outcome s produced by efficient Nash equilibria. Otherwise, the outcomes of the sim ulation more closely resemble the outcomes of inefficient Nash equilibria. (C) 2000 Elsevier Science B.V. All rights reserved. JEL classification: C7; C15; G34.