Credible communication and cooperation: Experimental evidence from multi-stage Games

• Published in: Journal of economic behavior & organization Vol. 81; no. 1; pp. 207 - 219
• Main Authors: Andersson, Ola, Wengström, Erik
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 2012
• Subjects: Communication; Cooperation; Economics; Economics and Business; Economics and Econometrics; Ekonomi och näringsliv; Experiments; Game theory; Messages; Nationalekonomi; Organizational Behavior and Human Resource Management; Renegotiation; Samhällsvetenskap; Social Sciences; Renegotiation; C92; Experiments; Cooperation; C72; Communication
• ISSN: 0167-2681; 1879-1751; 1879-1751
• DOI: 10.1016/j.jebo.2011.10.002

► This paper experimentally investigates cooperation and non-binding communication in a two-stage game. ► Two treatments are considered: one with only pre-play communication and one where subjects can also communicate intra-play between the stages of the game. ► Pre-play communication has a significantly lower impact on cooperation when intra-play communication is possible. ► The results suggest that the credibility of pre-play messages may depend crucially on future communication opportunities. It is well known that communication often serves as a facilitator for cooperation in static games. Yet, communication can serve entirely different purposes in dynamic settings as communication during the game may work as a means for renegotiation, potentially undermining the credibility of cooperative strategies. To explore this issue, this paper experimentally investigates cooperation and non-binding communication in a two-stage game. More specifically, two treatments are considered: one with only pre-play communication and one where subjects can also communicate intra-play between the stages of the game. The results highlight a nontrivial difference concerning the effects of pre-play communication between the two treatments. Sending or receiving pre-play messages has a positive and significant effect on cooperation if there is no possibility of intra-play communication. However, this effect is significantly reduced when when intra-play communication is allowed. The results suggest that the credibility of pre-play messages may depend crucially on future communication opportunities.