The category error: commons vs. open access
Garrett Hardin's "The Tragedy of the Commons" (Science, 1968)1 is built on a scenario: a pasture open to all herders. Each herder, acting rationally, will add one more animal because the benefit accrues to them while the cost of overgrazing is shared across all users. The logic iterates to collapse. Hardin concluded that shared governance of common-pool resources is structurally unstable. The only escapes are private property or state regulation.
This conclusion followed analytically from the model. The model's central assumption — a pasture "open to all," with no membership rules, no communication, no governance history — was never interrogated. Hardin labeled it "a commons." It isn't. It's an open-access regime, which is the opposite of a commons.
Open access: no defined boundaries, no membership, no rules for use, no exclusion mechanism. Any actor can extract any amount at any time. This is Hardin's pasture. It is correct that open-access resources face severe collective action problems.
Commons: bounded membership, defined rules for use, accountability mechanisms, and — critically — the governance capacity of the members themselves. The institutional content of a commons is precisely what Hardin's model assumes away.
The error is not minor. Privatization advocates spent decades using a model that assumes the absence of governance to argue against governance. Ostrom's first and most lasting contribution is demonstrating that the variable Hardin held constant — community governance capacity — is empirically the decisive one.
Hardin himself eventually acknowledged the error, in stages. In a 1991 book chapter he first reframed the argument around the "unmanaged" commons.3 A brief 1994 note in Trends in Ecology & Evolution used "unmanaged" in its title.4 In a 1998 Science essay he was explicit: the "weightiest mistake" in his original paper "was the omission of the modifying adjective 'unmanaged'" — and a managed commons "describes either socialism or the privatism of free enterprise. Either one may work; either one may fail: 'The devil is in the details.'"5 The policy apparatus built on the 1968 formulation did not adjust accordingly. Ostrom's work is, in part, the extended empirical brief for why it should have.
The empirical record: cases that shouldn't exist
If Hardin's model were correct, durable self-governing commons would be rare or absent from the historical record. They are neither. Ostrom and her colleagues at the Workshop in Political Theory and Policy Analysis at Indiana University documented dozens of cases across centuries and continents. Several have been operating continuously since the medieval period.
bylaws from 1483
The methodological challenge Ostrom faced was significant. These cases span radically different resource types, legal systems, cultures, and centuries. The standard social science move — abstract away the institutional detail to find the universal mechanism — would destroy the very variation she was trying to explain. Her response was to develop a framework capable of representing institutional complexity without flattening it. That framework is the IAD.
The IAD framework: what makes governance analyzable
The Institutional Analysis and Development framework is Ostrom's methodological contribution. It provides a common vocabulary for analyzing how rules, people, biophysical conditions, and community attributes interact to produce governance outcomes — without presupposing that any particular arrangement will succeed or fail.
The IAD framework's practical value is the rules-in-use concept. Most governance analysis works from formal rules — statutes, constitutions, contracts. Ostrom insisted on studying what rules actually govern behavior in practice. Communities develop extensive informal governance that is invisible to formal analysis but is doing the real work. The Swiss village that has been managing its alpine commons since the thirteenth century is not functioning on the basis of current Swiss property law. It has its own institutions, which predate the state and in many cases still operate alongside it.
This distinction matters for policy. Externally designed governance systems — including the privatization and nationalization that Hardin's model recommends — repeatedly fail not because they're wrong in principle but because they displace the informal institutions that were actually producing order, without replacing the function those institutions served. The wreckage of World Bank-mandated water privatization programs across the developing world in the 1990s is, in part, Hardin's model acting on Hardin's conclusions without Ostrom's corrective.
The eight design principles: annotated
Ostrom's eight design principles are the empirical core of Governing the Commons (Cambridge University Press, 1990).9 They are not prescriptions derived from theory. They are recurring patterns extracted from the cases — present in durable, successful commons and absent in failures. Ostrom herself noted that subsequent research has supported, qualified, and extended these principles but not refuted them as a class.
Polycentric governance: beyond the false binary
The deepest implication of Ostrom's work is not about fisheries or irrigation canals. It is about the architecture of governance itself. Hardin's framing produced a binary: private property or state regulation. These are described as the only two mechanisms capable of aligning individual incentives with collective welfare in the management of shared resources. Ostrom's evidence shows that both branches of this binary routinely fail, and that the cases of success involve a third option the model cannot see: polycentric self-governance.
The concept of polycentric order was developed by Vincent Ostrom, Elinor's husband and long-term collaborator, in a 1961 paper on metropolitan governance with Charles Tiebout and Robert Warren.7 The core claim: governance systems with multiple overlapping decision centers — each with bounded autonomy, none with monopoly authority, all subject to the same overarching rules — outperform monocentric systems on virtually every dimension when governing complex, spatially distributed problems.
Nationalization replaces governance with administration.
Polycentric governance is governance —
at the scale and density the problem requires.
The policy implications are specific. In a polycentric system, different governance centers handle different decision types at the scale appropriate to each. Local communities set use rules for the resource portion they know. Regional bodies coordinate across communities where resource flows cross local boundaries. National law sets the outer frame and adjudicates conflicts between levels. No level is sovereign over all decisions; each has delegated authority over appropriate ones.
The failure mode of centralization is epistemic: a single governance center lacks the local knowledge that distributed users have, systematically fails to adapt rules to conditions it cannot observe, and — critically — has no feedback mechanism capable of correcting its own errors at the speed that resource conditions change. The failure mode of pure privatization is different: it resolves the governance question by removing the resource from commons governance entirely, at the cost of concentrating surplus extraction in whoever holds the property right. Both moves are responses to the same problem — the difficulty of collective action. Both avoid rather than solve it.
Ostrom's contribution is to show that communities with sufficient autonomy, information, and time develop governance institutions that solve the collective action problem without either move. The design principles are a description of what those institutions have in common. They are an argument that the capacity is real, durable, and replicable — not utopian, not exceptional, not dependent on unusual conditions of human virtue. They are the product of ordinary communities doing what they need to do to survive.
Standard objections, addressed
Ostrom addressed this directly. The California groundwater basins in her 1990 study are contemporary, large-scale, legally complex, and polycentric in exactly the way her framework describes. The Maine lobster fisheries operate within and alongside modern US commercial law. Her later work with Charlotte Hess on knowledge commons — internet governance, open-source software, scientific research commons — applies the framework to large-scale, digital, non-rival resources.
The scaling objection assumes that complexity invalidates the framework. What Ostrom showed is that complexity requires polycentric governance — not that it makes commons governance impossible, but that monocentric governance (either state or market) fails at the complexity threshold where the locality assumption breaks down.
This was a serious methodological concern that Ostrom took seriously. Her response was experimental as well as observational. Beginning in the 1980s, she and her colleagues ran laboratory experiments on commons governance — controlled settings in which participants managed common-pool resources under varying institutional conditions. The findings consistently supported the observational results: when communication was permitted, when participants could develop their own rules, when monitoring was mutual rather than external, cooperation rates were dramatically higher than standard game-theoretic models predicted.
The causal mechanism she proposed — repeated interaction under conditions of mutual monitoring creates the conditions for graduated trust-building that sustains cooperation — is consistent with both the historical cases and the experimental results. It is also consistent with evolutionary theory of cooperation, which is why Kropotkin's empirical work and Ostrom's institutional analysis converge on the same conclusion through different routes.
The evidence here is genuinely mixed. Some studies find that social heterogeneity — inequality, ethnic diversity, conflicting interests — increases the transaction costs of commons governance and can undermine it. Ostrom acknowledged this and did not claim universal applicability. Her design principles specify community attributes as an input variable, not a constant.
The stronger version of the point: when commons governance fails in diverse communities, the failure is rarely endogenous to the community. It is almost always driven by external coercion — enclosure, colonization, legal override of community governance — that destroys the institutional conditions (P7: minimal recognition of rights) under which diverse communities can develop governance over time. The claim is not that diversity is irrelevant. It is that the relevant question is whether communities are given the conditions to develop governance, not whether homogeneity is a prerequisite.
This is the hardest case, and Ostrom was explicit that global commons governance is a distinct and harder problem. She did not claim that polycentric self-governance of local communities directly solves climate change. Her argument for global commons was more modest: polycentric governance systems — multiple overlapping institutions operating at different scales, including national and international levels — outperform the monocentric global treaty approach that has characterized international climate negotiations.
Her Nobel lecture — delivered December 2009, published as "Beyond Markets and States: Polycentric Governance of Complex Economic Systems" in the American Economic Review (2010)8 — explicitly extended the framework to the climate case. The argument: a polycentric approach allows experimentation, learning, and adjustment across many institutional contexts simultaneously. It builds social capital at local scales that changes the political conditions for higher-level action. It does not require universal agreement before action begins. The alternative — waiting for comprehensive global agreement — has a thirty-year track record of failure. The comparison class matters.