Evaluating others’ behaviour: a public-good experiment with ex-post communication

Abstract

The present study investigates the effect of ex-post communication on the voluntary provision of public goods. The experiment consists of two stages. In the first stage, paired subjects play a standard public-good game; in the second, each of them evaluates his or her partner’s contribution in the first stage and sends a free-form written message to him or her. The experimental results show that the mere presence of an opportunity for ex-post communication is not enough to promote cooperation. However, once subjects actually receive negative messages, they contribute significantly more in the following round. Positive messages, by contrast, do not have such an effect.

Keywords:

• public-good experiment
• ex-post communication
• evaluation
• cooperation

JEL Classification:

• C92
• H41

Acknowledgement

This research was supported by Grants-in-Aid for JSPS Fellows 211071 and 231657 from the Japan Society for the Promotion of Science.

Notes

¹ A large volume of the social psychology literature also confirms that preplay communication improves cooperation in prisoner’s dilemma games and n-person social dilemma games (e.g. Dawes, 1980; Sally, 1995).

² The experiment used the z-Tree programme (Fischbacher, 2007).

³ Subjects were informed in advance that a second stage would follow.

⁴ The backgrounds of the sampled students were diverse and included fields other than economics.

⁵ In the event of a tie, the investigator would have had the casting vote; however, there were no ties.

⁶ According to the Wilcoxon signed-rank test, this decrease was significant at the 10% level (z = 1.918), while the average contribution in EC did not significantly change at the 10% level (z = 0.752).

⁷ There was no collinearity between these independent variables.

⁸ Andreoni (1995) also noted that the decline in contributions might result from the frustration felt by cooperative subjects.