In economic relations and public goods provision, individual and public interests often clash because individuals have an incentive to "free ride" and not contribute. Researchers have conducted numerous experiments using the Public Goods Game (PGG) to study human behavior dynamics. Results from studies (Fehr and Gächter 2000, Nikiforakis 2008, Yamagishi 1988) confirm that cooperation among individuals cannot be sustained due to opportunistic behavior. To tackle this issue, a punishment mechanism was introduced in the experimental PGG, which has proven to be an essential tool in promoting cooperation. The punishment mechanism provides an incentive for opportunistic individuals to cooperate and, as a result, fosters a belief in cooperation among individuals.
Numerous studies have reported on the effectiveness of peer punishment in improving cooperation in PGG (Fehr and Gächter 2000, Herrmann, Thöni and Gächter 2008, Nikiforakis 2008). However, the effectiveness of promoting cooperation through peer punishment in PGG varies according to culture. For instance, several public goods game experiments with peer punishment have documented the existence of "antisocial" punishment, whereby sanctions are used against cooperators rather than free riders (Herrmann, Thöni and Gächter 2008). This paper focuses on whether decentralized (peer-to-peer) punishment mechanism can have an impact on prosocial behavior in PGG and existence of antisocial punishment in the Georgian context.
We performed the analysis based on the data retrieved from the experiment conducted by Mekvabishvili et al. (2022), which involved the standard ten-round PGG with peer punishment, where participants had the option to punish their group members. We analyzed the outcomes of the game and compared our results with the findings of similar experiments conducted in 16 different cities by Herrmann, Thöni, and Gächter (2008).
The results of the experiment conducted in Tbilisi suggest that the introduction of the peer punishment mechanism was an ineffective tool to promote cooperation, as the average contributions exhibited a declining pattern and remained well below 50% during all 10 rounds of the game. In the last period (round 10), the average contributions dropped to 21.4% of the total endowment, while 58.4% of subjects contributed zero to the public good. The average
contributions during the 10-round PGG amounted to 31.9% of the endowment. Compared to other cities, Tbilisi had the second-lowest average contributions after Athens (28.5%), and slightly lower than Istanbul (35.6%) and Riyadh (34.6%). All other cities showed significantly higher levels of contributions.
Turning to punishment behavior, the average punishment in Tbilisi was 4.8 points (out of 10 points), which was substantially higher than in experiments conducted in other cities. For example, the cities with the next highest average punishment points after Tbilisi were Muscat (1.3 points), Athens (1.0 points), and Riyadh (0.9 points).
We also investigated how individuals behaved when they punished other group members who contributed less, the same amount, or more than their own contribution to the public good. Our findings show that although free riders were punished strongly, there was still a significant amount of punishment directed towards those who contributed as much or more than the punisher. Antisocial punishment, but with less magnitude, was observed in other cities as well and was most pronounced in Muscat, Athens, Riyadh, Samara, Minsk, Istanbul, and Seoul (Herrmann, Thöni and Gächter 2008).
The comparison of the results of experiment conducted in Tbilisi with the results of Herrmann, Thöni, and Gächter (2008) revealed the significant role of the cultural factor in shaping economic relations. The experiment showed a large portion of antisocial punishment in the PGG, which ultimately led to a failure of economic cooperation: decentralized punishment incentives to cooperate had a negative impact on prosocial behavior in Georgia. Therefore, the decentralized or "peer punishment mechanism" was found to be an inefficient tool to promote cooperation in Georgia. The country has the lowest level of trust and cooperation norms among the countries included in the experiment. This is consistent with the high frequency of antisocial punishment observed in the experiment.
To summarize, the experiment showed that a significant amount of antisocial punishment occurred in the PGG, and the use of the peer punishment mechanism was not efficient to encourage cooperation. The results indicated that the decentralized punishment system, which aimed to motivate cooperation, had an adverse effect on prosocial behavior in Georgia. Our findings highlight the need to understand cultural factors in designing effective mechanisms to promote cooperation in economic interactions. However, it is important to consider that these findings were observed in an environment that lacked a suitable normative consensus and collective choice. Therefore, our findings call for more rigorous studies of the institutional environment that promote prosocial behavior.
The experimental designs and cooperation and peer punishment in PGG is a relatively new topic in Georgia, and few studies have been conducted on this issue. Future studies should
explore various institutions and settings. For instance, it would be interesting to compare the effectiveness of rewards and sanctions in Georgia. Moreover, it is worth exploring the effectiveness of centralized and decentralized mechanisms in promoting cooperation and prosocial behavior. The cultural and historical heritage that shapes an individual's social behavior is significant. Therefore, it would be interesting to repeat the experiment ineighboring countries of Georgia (e.g., countries in the South Caucasus) and conduct a comparative analysis of the results obtained.

Key words: Punishment of a partner (groupmate), decentralized punishment, the public goods game, cooperation, prosocial behavior.
JEL Codes: C90, C91, C93