Collective Action, Cooperation & The Commons
6.1 The Logic of Overuse
Whenever a resource is shared — a fishery, a pasture, the atmosphere — each user gains the full benefit of one more unit but bears only a fraction of the cost, spread across everyone. A herder adding one cow to a common pasture gets all the milk but shares the overgrazing. Each herder reasons identically. The pasture is destroyed.
An -player Prisoner's Dilemma: the dominant strategy is to take more, the Nash Equilibrium is depletion. Garrett Hardin named this "The Tragedy of the Commons" (1968), and his conclusion was stark: shared resources must be privatized or placed under government control. No third option.
He was wrong about that last part.
6.2 How Communities Actually Solve It
Elinor Ostrom spent decades studying communities that had governed shared resources successfully — some for centuries — without privatization and without state control. Irrigation systems in Spain. Fisheries in Maine. Forests in Japan and Nepal. Groundwater basins in Los Angeles. In each case, the users themselves had developed rules: who could use the resource, how much, when, and what happened to violators.
From these cases, Ostrom extracted eight design principles that distinguish commons that survive from commons that collapse:
- Clear boundaries. Both the resource and the group of users must be defined. If anyone can walk in, free-riding is uncontrollable.
- Rules matched to local conditions. The rules governing use must fit the specific resource — what works for an Alpine meadow won't work for a coastal fishery.
- Users make the rules. People who live under the rules participate in creating them. Externally imposed rules are ignored, resisted, or gamed.
- Monitoring. Someone watches whether the rules are followed. In successful commons, the monitors are often the users themselves.
- Graduated sanctions. First offense gets a warning. Repeated violations escalate. This balances forgiveness (people make honest mistakes) with deterrence (persistent cheaters are punished).
- Accessible conflict resolution. Cheap, quick ways to resolve disputes — not distant courts or expensive lawyers.
- The right to self-govern. External authorities (governments, corporations) must not undermine the community's ability to manage its own commons.
- Nested layers for large-scale commons. When the resource is too big for one community, governance should nest: local rules within regional rules within national frameworks.
Empirically validated across hundreds of case studies on every continent. Linux and Wikipedia — volunteer contributors, no employer, no coercive authority — are functioning commons at massive scale, and they meet Ostrom's conditions: defined communities, monitoring through version control and edit histories, graduated sanctions (reverts, bans), and collective-choice governance.
For decades the policy answer to shared resources was binary: privatize or regulate. Ostrom demonstrated a third option — community self-governance — that often works better, because local users hold detailed knowledge that neither markets nor distant regulators possess.
6.3 The Shadow of the Future
We previously analyzed the Prisoner's Dilemma as a one-shot game. But most real interactions are repeated — the same players meet again and again.
This changes everything. In a repeated game, you can punish defection in future rounds. The threat of future punishment makes cooperation rational now — but only if the future matters enough. The critical variable is the discount factor : how much players value the future relative to the present. High (patient players, long-term relationships) sustains cooperation. Low (impatient players, approaching end) restores defection.
Robert Axelrod tested this with tournaments where submitted strategies competed in repeated Prisoner's Dilemmas. The winner was tit-for-tat: cooperate on the first move, then mirror whatever the other player did last. It succeeds because it is nice (never cheats first), retaliatory (punishes cheating immediately), forgiving (returns to cooperation as soon as the other player does), and transparent (easy to understand and predict). A later refinement — generous tit-for-tat, which occasionally forgives defection randomly — performed even better by preventing the death spirals of mutual retaliation that small mistakes can trigger.
Cooperation is easier when the future is valued, cheating is detectable, and retaliation is credible. Anything extending the time horizon (stronger relationships, longer-term contracts, reputational stakes) helps. Anything shortening it (anonymous one-shot deals, approaching deadlines, lame-duck governments) hurts.
This is why international climate cooperation is so hard. The interactions are repeated, but the shadow of the future is weakened by political cycles (governments change every 4–8 years), discount rates (politicians care about the next election, not the next century), and imperfect detection (emissions are hard to verify). The Kyoto Protocol (1997) attempted binding targets; the US never ratified, Canada withdrew, and most signatories missed their goals. The Paris Agreement (2015) switched to voluntary pledges — a weaker commitment, but one more countries were willing to make. Whether Paris sustains cooperation depends on whether the repeated-game conditions hold.
6.4 Why Concentrated Interests Win
There is another reason climate cooperation fails, and it explains one of the most persistent patterns in democratic politics.
A steel tariff costs each American consumer a few cents per product. It benefits a few hundred steel executives and thousands of steelworkers enormously. The tariff is inefficient — total cost to consumers exceeds total benefit to producers. Yet it persists. Why?
Because collective action is a public good with a free-rider problem. If consumers organize to fight the tariff, every consumer benefits — including the millions who contributed nothing. If everyone reasons this way, no one organizes. But the steel industry is a small group with concentrated per-capita stakes — each firm's benefit justifies lobbying. The millions of consumers are a large group with diffuse stakes. No one consumer's loss is worth organizing over.
This asymmetry — small groups with concentrated interests organize; large groups with diffuse interests don't — is the core of Mancur Olson's Logic of Collective Action (1965), and it predicts democratic politics with uncomfortable accuracy. Agricultural subsidies persist because a few thousand large farmers gain enormously and lobby effectively, while hundreds of millions of consumers and taxpayers each lose a few dollars and never notice. Pharmaceutical patent extensions persist because a handful of companies invest millions in lobbying, while every patient who pays inflated drug prices bears a cost too small to organize around. Regulatory capture — where the agency regulating an industry ends up protecting it — is an Olson problem: the regulated industry has concentrated stakes in influencing the regulator; the diffuse public does not.
The implication for democratic theory is grim. Democracy is supposed to represent the majority. Olson's logic predicts that well-organized minorities will systematically prevail over disorganized majorities.
6.5 When Does Collective Action Work?
The temptation is to see contradiction — Ostrom says cooperation works, Olson says it doesn't. The precise reading: they explain different conditions.
Olson explains why collective action fails by default in large, anonymous groups without institutional structures to overcome free-riding. Ostrom explains the institutional conditions under which it succeeds: clear boundaries, monitoring, graduated sanctions, local rule-making.
The synthesis: collective action succeeds or fails depending on whether the institutional conditions are present. Where they are (Ostrom's cases, open-source software, local fisheries), cooperation is robust. Where they are not (global climate negotiations, diffuse consumer interests, large anonymous populations), Olson's logic dominates.
The policy task is to build institutions that satisfy Ostrom's conditions — not to hope that goodwill overcomes the underlying incentives.