Contribution
A synthesis lecture. Its contribution is the argument that the mid-century framework of two organizational forms (market and state), two kinds of good and one model of the individual does not fit the evidence. It offers a polycentric, multi-scale account backed by meta-analysis, experiments and field studies.
Key insight
Successful self-governance is not explained by particular rules, which vary endlessly across sites. It is explained by structural regularities: users who make and monitor their own rules, sanctions that start low, cheap local dispute resolution, and governance nested in layers when the resource is embedded in something larger.
Assumptions
- The design principles are inductive regularities from coded cases, not derived or deliberately designed: “I did not mean that the fishers, irrigators, pastoralists, and others overtly had these principles in their minds” (p. 422).
- Individuals are boundedly rational, learn in repeated settings with feedback, use heuristics, and hold norms (§7A).
Key results
- Policing (§2A): “most efficient producers supply more output for given inputs in high multiplicity metropolitan areas” (quoting E. Ostrom & Parks 1999).
- Design principles (§4E), updated from Cox, Arnold & Villamayor-Tomás (2009): 1A user and 1B resource boundaries; 2A congruence with local conditions, 2B costs proportional to benefits; 3 collective-choice arrangements; 4A monitoring users, 4B monitoring the resource; 5 graduated sanctions; 6 conflict-resolution mechanisms; 7 minimal recognition of rights; 8 nested enterprises. About two thirds of over 100 studies confirm them.
- Lab (§5A): baseline overuse beyond Nash; face-to-face communication raises returns; imposed costly sanctioning lowers net returns; self-chosen sanctioning reaches 90% of optimum.
- Field experiment (§5B): an imposed rule enforced with 1/16 inspection probability raised extraction relative to communication alone.
- Nepal irrigation (§6A): farmer-governed systems better on physical condition, tail-end water and productivity; confirmed on 229 systems.
- Forests (§6B): no difference in density between protected and other forests (163 forests); monitoring correlates with density across 178 user groups.
Claims
| id | claim | strength | support |
|---|---|---|---|
| C1 | Many independent providers can be more efficient than consolidation | moderate | 1970s police studies, 80 metropolitan areas |
| C2 | Long-surviving CPR institutions share the design principles | moderate | coded cases; two-thirds confirmation in a meta-review |
| C3 | Communication and self-chosen sanctions sustain cooperation | strong | repeated lab results, consistent with a wider literature |
| C4 | External regulation can crowd out voluntary cooperation | moderate | one field experiment plus cited studies |
| C5 | Local monitoring, not formal protection, predicts forest condition | moderate | correlational field data, with controls |
Concepts
- polycentric — many formally independent centres of decision; whether they form a system is empirical.
- common-pool resource — a good from which exclusion is costly and use by one subtracts from others.
- design principle — a structural regularity of long-enduring institutions; Ostrom suggests “best practices” might have been clearer (n. 5).
Connections
Builds on V. Ostrom, Tiebout & Warren (1961) for polycentricity, and on the IAD framework and the case meta-analyses behind Governing the Commons (LIT-737). It argues against Hardin’s tragedy as a general model.
Bearing on the record
Source of THEORY-149. The owner’s essay (§8.2) cites Ostrom for “how distributed monitoring and authority can contribute to institutional regulation”. Principles 4 and 8, and the forest monitoring results, are the specific support. Two cautions follow from the lecture itself. First, the principles are about users governing a resource they share, and a firm’s governance of itself is a different structure. Second, Ostrom’s lesson is that fit and legitimacy matter more than form, so the essay should not read polycentricity as good in itself.
Limitations
A lecture: the evidence is summarised and cited, not presented. The design principles are correlates of survival, not tested causes. Several field results are correlational.
Open questions
Which principles are causally necessary, and how they apply outside natural-resource settings. The lecture flags the link between micro and contextual variables as the main open task (§7D).