Public-goods game

The Public-goods game is an experimental model of a basic collective dilemma: each participant can benefit from a shared pool, but has an individual incentive to keep resources rather than contribute. If everyone contributes, the group can do substantially better than if everyone keeps their resources; if enough people free ride, however, the common benefit shrinks. The game makes visible how cooperation depends not only on a calculation of private gain, but also on expectations about others and on whether free riding can be checked.

Structure and Incentives

In the version described by economists Ernst Fehr and Simon Gächter, four people play for four rounds, each beginning with twenty tokens. On each round, a player may contribute tokens to a public pot or keep them. A contributed token costs its donor one token, while returning 0.4 tokens to each of the four players: the contributor personally receives only 0.4, but the group gains 1.6 in total. If nobody contributes, each player ends with twenty tokens; if everyone contributes all their tokens, each ends with thirty-two. The incentives therefore point in opposite directions. The group benefits from contributions, but a player can do best individually by withholding tokens and benefiting from what others put in.

This tension is why the game serves as a standard behavioral-economics experiment alongside the ultimatum and dictator games. Those games can be played according to a narrowly self-protective strategy, forgoing possible mutual gains, or in a more prosocial way. The public-goods game isolates the cost of contributing, the shared return, and the temptation to free ride: its central question is not whether cooperation benefits the group, but how it can persist when any one player can benefit without paying its cost.

Cooperation and Its Erosion

Across public-goods experiments played throughout the developed world, most participants do not begin by behaving as pure free riders. They initially contribute about half their tokens to the common pot. But contributions tend to decline as rounds pass and players see others withholding resources. By the end, 70 to 80 percent of players are free riding, leaving the group poorer than it could have been.

Fehr and Gächter’s account distinguishes three broad patterns of behavior. Roughly a quarter of participants are consistently selfish in the experiment’s economic sense: they always free ride. A small minority are altruists who contribute heavily from the outset and continue even when others do not reciprocate. The largest group consists of conditional consenters. They begin by contributing at least some of their resources, but observing free riding makes them less willing to continue. In the ordinary game, cooperation is therefore vulnerable not only to the players who never contribute, but also to the discouragement of those whose participation depends on seeing others do their share.

Visibility, Punishment, and Reciprocity

Fehr and Gächter tested how the rules might protect cooperation by changing what players knew and could do. At the end of each round, they revealed each person’s contribution, making free riders visible to the group. Players could then spend a third of a token to take one token away from a free rider. The penalty cost the person imposing it, so punishing another player was economically irrational in the narrow sense of maximizing one’s own immediate token total.

Players nevertheless spent resources to punish free riders, and the free riders began contributing their fair share. The effect extended beyond rounds in which punishment could be imposed: in the final rounds, when no further punishment was possible, participants continued contributing. The experiment thus suggests that costly sanctions can do more than deter a particular act while the threat remains in place. They can help sustain expectations of reciprocity, allowing people to keep cooperating without continually using the sanction itself.

From Experimental Rules to Collective Life

The game’s logic extends beyond the laboratory. Taxation, for example, finances services that people can benefit from whether or not they personally contribute. Roads, education, public protection, and scientific research depend on shared contributions, while the possibility of enjoying these benefits without paying creates a free-rider incentive. A functioning system therefore requires more than a formal rule: people must have reason to believe that others are participating and that those who do not contribute may be caught and punished.

This account also connects the game to evidence from small-scale societies. In research supervised by Samuel Bowles, twelve field researchers—eleven anthropologists and one economist—studied fifteen societies with different forms of subsistence, including foraging, slash-and-burn cultivation, nomadic herding, and small-scale agriculture. Across the societies, people departed significantly from purely self-interested strategies, though the pattern varied between cultures. Greater integration with markets was associated with greater prosociality: participants from more market-oriented societies made higher offers in the dictator and ultimatum games, cooperated more in public-goods games, and showed strong reciprocity. The comparison places the experimental dilemma within a wider account of how cooperation and expectations of reciprocity take shape.

The Public-goods game does not imply that collective cooperation follows automatically from shared advantage. Its rounds show the opposite: contributions can unravel when participants see others free riding, while visible conduct and a credible response to defection can help restore them. Cooperation is sustained not by assuming that everyone will contribute from benevolence, but by conditions that keep conditional participants from feeling that they alone are paying for the common good.