Search

Central sanctioning institution controlled by third party is a double-edged sword for the evolution of cooperation. On the one hand, the institution can protect cooperators from the exploitation of free-riders; on the other hand, a corrupt third party can cause the institution to fail, thus resulting in the collapse of cooperation. However, previous relevant studies have always explored that third parties are bribed by free-riders, and therefore forgo sanctioning free-riders. Although bribery is an important component of corruption, it is not comprehensive. In this paper, we consider third parties who embezzle public funds, which is characterized by the fact that the deterrence of sanctions remains even if corruption occurs. There is a new dilemma here, where cooperators are exploited by third parties when they are in the institution and by free-riders when out of the institution. An evolutionary game model is established to represent this dilemma, and a complex network model is constructed to represent interactions between third parties and players. We discuss the evolutionary dynamics of cooperation, trust and corrupt behaviours both in finite and infinite population, while find that players trust corrupt third parties and maintain cooperation as long as the sanction institution is sufficiently effective and the corruption rate is not very high. This work presents a novel framework for representing interactions between players and third parties, and suggests that exploitation by free-riders is less acceptable to cooperators than exploitation by third parties. [ABSTRACT FROM AUTHOR]