Public Goods Game¶
Model voluntary contributions to a shared account whose private and group returns pull in different directions.
Core Idea¶
The linear public-goods game gives each of \(n\) players an endowment \(e_i\) and a voluntary contribution \(c_i\) to a common account. A baseline material payoff is \(\pi_i=e_i-c_i+\alpha\sum_jc_j\): each keeps what they do not contribute and receives the same per-capita return \(\alpha\) on all contributions. When \(1/n<\alpha<1\), an extra unit lowers the contributor's own material payoff by \(1-\alpha\) while raising the group's total by \(n\alpha-1\). Thus zero contribution is individually dominant for own-payoff maximizers in this one-stage baseline, while full feasible contribution maximizes aggregate baseline payoff. With an equally divided multiplied pot, \(\alpha=r/n\) and the condition is $1<r<n$. Actual participants need not choose the monetary best response.[ref-bf7ebc22447c][ref-ce177a0c2225]
Scope of Application¶
This is a reusable experimental-economics game form, not a real public good itself. Fischbacher, Gächter and Fehr used a one-shot four-person game with 20 tokens each and \(\alpha=0.4\), eliciting both an unconditional contribution and a contingent contribution schedule. Fehr and Gächter used the same four-person, 20-token, \(\alpha=0.4\) first-stage payoff over ten periods with fixed or re-matched groups, with or without an additional costly peer-punishment stage. These are unlike elicitation/institutional variants of a shared contribution skeleton; punishment is not required by the baseline definition.[ref-bf7ebc22447c][ref-ce177a0c2225]
Clarity¶
State players, endowments, feasible choices, common account, \(\alpha\) and the payoff rule before discussing equilibrium. The private effect is \(-1+\alpha\); the group effect is \(-1+n\alpha\). The result depends on the stipulated monetary objective and parameter region, not on a claim that everyone is selfish. A one-shot conditional contribution schedule does not mean players observed others before an ordinary simultaneous move. Adding a punishment stage changes the extended game's payoffs. The live Public Goods classifies a good by consumption properties; it is not identical to this action-and-payoff model.[ref-bf7ebc22447c][ref-ce177a0c2225]
Manages Complexity¶
One equation isolates the private opportunity cost, common return and group gain that together generate the free-rider dilemma. It gives a sharp benchmark for interpreting observations: in Fischbacher and colleagues' sample, 22 of 44 subjects' schedules were classified conditionally cooperative and 13 pure free-riding. Fehr and Gächter's partner groups had final-period mean contributions a little above 3 tokens without punishment versus above 18 with punishment. These are study-specific observations; a higher contribution need not imply higher net payoff after sanction costs, and neither study establishes a universal human response.[ref-bf7ebc22447c][ref-ce177a0c2225]
Abstract Reasoning¶
First derive the signs of \(-1+\alpha\) and \(-1+n\alpha\) for the actual parameters. At \(\alpha\geq1\), contributing is not privately costly in the baseline formula; at \(n\alpha\leq1\), it is not aggregate-payoff-improving. Then identify whether an experiment tests simultaneous choice, conditional strategy elicitation, repetition or punishment. Do not infer a single motive from contribution alone, and do not carry the baseline best-response proof into an extended game without reanalyzing its actions and payoffs.[ref-bf7ebc22447c][ref-ce177a0c2225]
Knowledge Transfer¶
Transfer the players–endowment–contribution–common-return roles to another formal game only when its choices and payoffs are stated. The \(r/n\) shorthand and incentive wedge do not literally map onto every real shared-benefit phenomenon; the seed's microbial analogy is therefore omitted. Independent identity and DAG review remain pending.[ref-bf7ebc22447c][ref-ce177a0c2225]
[^ref-bf7ebc22447c]: Urs Fischbacher, Simon Gächter and Ernst Fehr, “Are people conditionally cooperative? Evidence from a public goods experiment”, Economics Letters 71 (2001), 397–404, §2 Eq. 1 and design, §3 results. [^ref-ce177a0c2225]: Ernst Fehr and Simon Gächter, “Cooperation and Punishment in Public Goods Experiments”, American Economic Review 90(4) (2000), 980–994, §I Eq. 1/Table 1 and §III Result 5/Table 4.
Neighborhood in Abstraction Space¶
Public Goods Game sits in a moderately populated region (54th percentile for distinctiveness): it has near-neighbors but no dense thicket of look-alikes.
Family — Strategic Games & Equilibrium Concepts (13 abstractions)
Nearest neighbors
- Ultimatum Game — 0.86
- Rubinstein bargaining model — 0.86
- Volunteer's Dilemma — 0.86
- Folk Theorem (Repeated Games) — 0.85
- Portfolio Optimization — 0.85
Computed from structural-signature embeddings · 2026-10-08