Rational and Cooperative Strategies in an Arms Race. Conclusions from Empirical Research

Abstract

Goal – this article describes a proposed model of arms race game and the results of its application in an experiment measuring the inclination of human players to participate in a non-productive competition such as an arms race. Research methodology – methods in this research include: designing a game of an arms race based on a prisoner’s dilemma and one dollar auction, theoretical analysis of strategies to determine the best rational strategy for the player, conducting the experiment in various conditions and comparison between theoretical and empirical results. Score/results – the analysis of various strategies showed that the most rational strategy is the minimum expenditure on armaments to allow effective defense against an adversary. The empirical results showed that players spent significantly higher amounts on armaments than theoretical predictions would have assumed, and 15 out of 26 games did not end close to Nash Equilibrum, which showed that the participants did not play according to rational calculation. The ability to communicate effectively resulted in players being more cooperative. In addition, those who said they played computer games performed better than the others, and finance and accounting students were more likely to choose strategies based on competition. The research supports the thesis that people are inclined to compete even when the competition brings them losses. Originality/value – the proposed game of an arms race is a new model, that can be used to simulate an arms race and to measure rationality of human players. The conducted experiments also provide evidence about human behavior regarding choosing cooperation or competition.