Decentralization Papers V. Ostrom’s Answer: Neither State nor Market

Decentralization is not an ideology and belongs to no party. It is a design constraint. And Elinor Ostrom proved, against the textbooks, that ordinary communities can govern shared resources without a boss.

Ask which ideology decentralization belongs to and every camp raises a hand. Anarchists claim it: from Proudhon's mutualism to Kropotkin's mutual aid, theirs is the oldest sustained argument that free association can do what command structures do, without the command. Classical liberals claim it: Hayek's case against planning is a decentralist argument, and federalism is decentralization written into constitutional law. Catholic social teaching contributed the principle of subsidiarity (no higher body should absorb what a lower one can do), which now sits, at least on paper, in the treaties of the European Union. And the cooperative and communal wing of socialism claims it, from the Paris Commune to the worker-owned firms quietly operating on every continent. When every rival tradition holds a piece of an idea, the idea is not a member of any of them. Decentralization is not an ideology. It is a constraint you can impose on any of them, and a test most of their implementations fail.

For much of the twentieth century, official social science barely admitted a third option existed. The textbook framing was a two-slot machine: resources are governed by the state or by private owners, and anything shared by a community is doomed. That is the famous tragedy of the commons, canonized in Garrett Hardin's 1968 essay, in which every herder overgrazes the shared pasture because restraint by one is subsidy to the rest. The metaphor hardened into policy: nationalize the commons or parcel it out, but never leave it to its users.

Elinor Ostrom went and looked. Across decades of field work (Swiss alpine pastures, Japanese village forests, Spanish irrigation courts, Filipino zanjera canals, fisheries, groundwater basins), she and her collaborators documented communities governing shared resources successfully for generations, sometimes centuries, with neither state ownership nor privatization. The tragedy was real but not fated: it afflicted commons whose users could not make and enforce their own rules. Where they could, they built institutions (boundary rules, monitoring, graduated sanctions, cheap arbitration) that often matched or outperformed both slots of the machine. Governing the Commons, published in 1990, assembled the evidence; in 2009 it earned her the Nobel memorial prize in economics, the first awarded to a woman.

What makes Ostrom this series' hinge is that she turned decentralization from a slogan into an engineering discipline. Her design principles read like a specification for power that stays distributed: boundaries clearly defined; rules fitted to local conditions and modifiable by those they bind; monitors accountable to the users; sanctions graduated rather than ruinous; conflict resolution cheap and near; the right to self-organize recognized rather than licensed; and (for commons larger than one community) governance in *nested, polycentric* layers, each scale handling what only it can. Compare that list with Part I's foragers and their watchful coalitions, or Part IV's Commune with its delegates recallable at any moment: the same architecture, discovered independently wherever power is successfully kept plural.

Polycentricity is the deep answer to the strongest objection centralizers have: that someone must govern the whole. Ostrom's reply: many centers, overlapping, none sovereign over all. A watershed does not need a single owner; it needs the irrigators' court, the village forest council, the basin authority, and the national law of contracts, each constrained by the others. Redundant, messy, and robust, where the single center is tidy, legible, and brittle. Part II's granary keeper and Part IV's Gosplan are the same failure at different scales; polycentric order is the alternative that does not simply relocate the throne.

Ostrom's honesty about limits is part of her authority. Self-governance is not automatic; it fails where boundaries blur, where users cannot see each other's behavior, where higher powers refuse communities the right to make rules at all. Her principles are conditions, not miracles. But that is exactly what makes them usable. They tell you what a decentralized institution must supply: visible behavior, cheap verification, rules the governed can change, exit that is real. Hold that list. It is about to become a checklist for software.

Because here is the closing thought of this part: everything Ostrom's principles require (visible behavior, verification any member can perform, rules the governed can change, exit that is real) is precisely what the internet promised, failed to deliver, and a newer generation of protocols now claims to provide again. Whether that claim survives contact with reality (and what it would take for it to) is Part VI.