Public goods game

from Wikipedia, the free encyclopedia

The public goods game , also public goods game or public goods game , or ÖGS for short, is part of game theory and is the subject of experimental economics . It is used as a model to analyze the provision of public goods . This analysis is important because, in contrast to private goods, public goods can be consumed multiple times or at the same time, but nobody can be excluded from consumption. Because nobody can be excluded, it is rational not to bear any costs for the provision ( free rider problem ).

Game flow

In the standard version of the game, the participants decide in secret how much of their initial equipment they want to invest in a public good. This pot is slightly multiplied at the end and then divided equally between all players. So players who have done nothing to provide the good also get their share. The game win of a player results from the public payment and the retained initial equipment. After the end of the game, this profit was usually converted into real currency at a known exchange rate. B. Study achievements swapped.

Multiplication factor

In the standard variant, the multiplication factor for the public good is between 1 and the number of players. A factor equal to one means that every player gets exactly his contribution back if everyone makes the same contribution. In this case the players would be indifferent between the public good and the simple withholding of all their wealth. A factor equal to the number of players means that the payout ratio and the multiplication ratio are the same. A higher factor would therefore always lead to a profit for all depositors.

Regardless of the size of the group and return flow of contributions, the group as a whole is in the best position when all participants contribute all their private equipment to the public good. This result would be the Pareto optimum : If each player pays in his entire initial equipment from the start, the group receives the maximum possible payout from the game master.

Expected and actual results

Standard economic theory, using the Nash equilibrium, predicted that no player would make group contributions when the game is finite. This dominant strategy of each player applies regardless of how each of the other players behaves. Since it is rational not to cooperate in the last round of finally repeated games, it is also rational not to contribute in any round because of the backward induction . This free rider problem leads to the tragedy of the commons : Although everyone in the group wants to have and use the public good, nobody pays for it (voluntarily). As a result, there is an undersupply of the public good compared to the state of the Pareto optimum. The Nash equilibrium also predicts that larger groups and / or lower returns (due to a lower multiplication of the public good) led to less cooperation because the incentives to provide the public good continued to decline.

In fact, the undersupply associated with the Nash equilibrium rarely or not at all occurs in experiments; those involved tend to invest at least a small part of their private equipment. The extent to which individual individuals participate varies greatly. However, the social optimum (complete cooperation of all participants) is rarely or never achieved, but more often in non-repeated games. Contrary to widespread opinion and economic theory, even a larger group or a higher multiplication factor does not necessarily lead to more cooperation or higher individual contributions to the public good. The sociological interpretation of these findings emphasizes group cohesion and cultural norms to explain the prosocial outcomes of public-goods games.

Explanatory approaches

After the first public goods games, the interest of science focused on why the game participants deviate from economic rationality or whether the theory is more likely to show inadequacies. Practically oriented studies also investigated inducing group cooperation with a suitable incentive system. This can be used to derive indications for many problem solutions in society (see section Modifications ).

Although the discussion about the results is not over, some explanations emerged early on:

  • Among other things, the game-theoretic Nash equilibrium required complete information about returns from the public good as well as the amount of initial equipment of all players. However, modifications with more and less information show no difference in behavior.
  • Since a higher level of cooperation was observed in individual games than in repeated games, the learning hypothesis was formed: At least some players must first learn how to behave rationally in optimization problems and develop their strategy. A time-delayed adjustment to the dominant strategy was also observed in other games. The learning hypothesis could neither be verified in public goods games nor in other games.
  • Players do not act according to the logic of backward induction in repeated games, that is, they do not think from the end in multi-round games.
  • Game theory is based on the fact that players maximize their utility exclusively through monetary returns. Some economists turned therefore one that theory to grab short, and players their benefits maximize by other accompaniments: So players would also benefit from the "warm glow" ( english hot glow ) of giving pull, or incorporate fairness calculi in their decisions .

criticism

However, it was also criticized to what extent simple, abstract, short and finite laboratory experiments in manageable groups are suitable for making predictions for human behavior in complex everyday situations. In reality, it cannot be ruled out that market players will meet again and then possibly feel the consequences of their actions from previous actions - such as non-cooperation.

Modifications

Repeated games

Repeated games mean that the same game is repeated in the same or a different composition for a certain number of rounds. A typical result is a decreasing share of contributions to the public good compared to a one-off game. Players tend to contribute less themselves than in the preliminary round when they see others giving less too. The effect repeats itself in the next round, but now from a lower base. However, nothing is never contributed by everyone, because a hard core of “donors” remains.

One explanation for this is the inequality aversion ( English inequity aversion ). Repeated play gives players an opportunity to build reputation as observed by others. On the other hand, players get to know certain characteristics of the entire group and continuously adapt their behavior. More precisely: They adjust their expectations about future payments from the other group members, because nobody wants to contribute to the public good alone. If players observe that other players receive a larger repayment for a smaller contribution, this contradicts their sense of fairness.

Transparent / Open Games

If the contribution amount or even the identity of the players is made transparent, individual contributions are regularly higher. This result applies regardless of the specific experiment setup, i.e. whether the players are known from the beginning, only in pairs, or are named at the end of the experiment, etc.

Benefit events regularly make this fact their own, where the donors are regularly named and some are honored according to their donation amount.

Reward and / or Punishment

The use of rewards or punishment has been the subject of much study. As a rule, the players punish each other after the public good has been provided. A player is chosen at random who punishes a group member at his own expense. Rewards work in the same way. Alternatively, rewards or penalties are automatically carried out by the game master according to a known rule. A key finding is that rewards and penalties are used as different means: rewards are not the same as non-penalizing, while a penalty is not considered an absent reward.

Penalties are carried out even at a lower cost and in most experiments lead to higher contributions or higher cooperation. The effect of rewards alone, on the other hand, is weaker. This must not be confused with higher group payouts: Since penalties cost money, increased cooperation does not necessarily mean higher payouts from the public pot. At least in the first round, penalties can lead to (marginally) lower group payouts.

Many studies therefore highlight the combination of punishments and rewards. It leads to both increased cooperation and higher contributions. This applies to repeated games in changing groups as well as to identical groups.

Asymmetric costs and / or benefits

Asymmetry in terms of the private costs of provision or the payout from the public pot have a direct impact on player behavior. They react more strongly to monetary incentives and behave rationally according to economic theory. However, even here more is contributed to the public good than in the Nash equilibrium .

Framing

Different representations of the same structure or the same game ( English framing ) show behavior that differs from the original game . A variant of framing is the association with real problems in which public goods have to be provided. These can be climate protection negotiations, the construction of a road or souvenirs for a private celebration. This allows the players to draw information about the preferences of the other players, assess the likelihood of their actions and perceptions.

The effect of an association (attribute framing) is different and depends on the personal experiences of the players. This is especially true for one-time games, where players (can) predict the behavior of other players only on the basis of their own experience in reality. Even players of the same culture can have different terms of the same attribute and respond with both higher and lower contributions.

In addition, every game can be presented as a choice between winning or a choice between losses. Because of the framing effect , players react completely differently: If public good games are presented as a loss instead of a profit (i.e., a contribution to a private good reduces other players' payouts) the contributions are significantly smaller.

See also

Individual evidence

  1. ^ A b Andreoni, James : Why free ride? Strategies and learning in public goods experiments . In: Journal of Public Economics . tape 37 , 1988, pp. 291-304 .
  2. ^ A b Isaac, R. Mark, James M. Walker, and Arlington W. Williams: Group Size and the Voluntary Provision of Public Goods: Experimental Evidence Utilizing Large Groups . In: Journal of Public Economics . tape 54 , no. 1 , 1994.
  3. ^ Kagel, J. and D. Levin: Independent private value auctions: Bidder behavior in first, second, and third price auctions with varying numbers of bidders . In: Economic Journal . tape 103 , 1993, pp. 868-879 .
  4. ^ Andreoni, James: Giving with impure altruism: Applications to charity and Ricardian equivalence . In: Journal of Political Economy . tape 97 , 1989, pp. 1447-1458 .
  5. ^ Kahneman, Daniel, Jack L. Knetsch, and Richard H. Thaler ,: Fairness and the assumptions of economics . In: Journal of Business . tape 59 , 1986, pp. 285-300 .
  6. ^ A b Levitt, Steven D. and John A. List: What Do Laboratory Experiments Measuring Social Preferences Reveal about the Real World? In: The Journal of Economic Perspectives . tape 21 , no. 7 , 2007, p. 153-174 .
  7. ^ A b McGinty, Matthew, and Garrett Milam: Public Goods Contribution by Asymmetric Agents: Experimental Evidence . In: Social Choice and Welfare . 2012.
  8. ^ Fehr, E. and KM Schmidt: A Theory of Fairness, Competition, and Cooperation . In: Quarterly Journal of Economics . tape 114 , 1999, pp. 817-868 .
  9. ^ Rege, Mari and Kjetil Telle: The impact of social approval and framing on cooperation in public good situations . In: Journal of Public Economics . tape 88 , no. 7–8 , 2004, pp. 1625-1644 .
  10. ^ A b Andreoni, James, William Harbaugh and Lise Vesterlund: The Carrot or the Stick: Rewards, Punishments, and Cooperation . In: The American Economic Review . tape 93 , no. 3 , 2003, p. 893-902 .
  11. ^ A b Sefton, M., R. Shupp and JM Walker: The Effect of Rewards and Sanctions in Provision of Public Goods . In: Economic Inquiry . tape 45 , no. 4 , 2007, p. 671-690 .
  12. ↑ a b Edge, David G., Anna Dreber, gates Ellingsen, Drew Fudenberg and Martin Nowak: Positive Interactions Promote Public Cooperation . In: Science . tape 325 , 2009, pp. 1272-1275 .
  13. ↑ Dufwenberg, Martin, Simon Gächter and Heike Hennig-Schmidt: The framing of games and the psychology of play . In: Games and Economic Behavior . tape 73 , no. 2 , 2011, p. 459-478 .
  14. ↑ Willinger, Marc and Antohny Ziegelmeyer: Framing and cooperation in public good games: solution to experiment with an interior . In: Economics Letters . tape 65 , no. 3 , 1999, p. 323-328 .