Thoughts on Polycentrism: a proposition and two open questions

I’m a huge fan of polycentrism (radical decentralization, if you prefer). I find the arguments for institutional trial & error as theoretically appealing as they are empirically supportable. And I believe we have a great on-going conversation about the theoretical underpinnings and practical implications of these ideas. There are, however, under explored areas within our discussion which should be further fleshed out. And it’s my hope that in highlighting these areas, we can move the discourse forward.

The Goldilocks Problem

One unexplored challenge is the task of defining appropriate scale. We often talk in terms of decentralization as if there’s no limit to how far we should disaggregate institutions, but the reductio ad absurdum is a world of autarkic individuals. What’s needed here is an account of decentralization’s lower bound and for that we can turn to Ronald Coase.

Coase’s The Nature of the Firm explained that we establish firms when transaction costs make decentralized coordination via price signals too expensive. While this work was written in the context of for-profit firms, we can apply the same logic to almost any kind of human organization. The entities we think of as governance systems exist to help us confront collective action problems. These collective action problems still pertain to exorbitant search and bargaining costs, they’re just not explicitly linked to costs associated with commercial transactions. In recognizing this we find our lower bound.

Decentralization should be taken to the point at which the marginal cost of additional decentralization exceeds the marginal benefit due to rising transaction costs. And on the opposite end, aggregation or institutional centralization should stop before hitting the diseconomies of scale with which we’re all already familiar.

I believe it is important to frame our thinking in terms of seeking a balance between these two extremes. Doing so allows us to paint a clearer theoretical picture as well as better respond to on-the-ground realities when discussing practical reform. Having lived in the San Francisco Bay Area for nearly a decade, I’ve seen the downside of too much decentralization. There’s a lot of history and context that goes into explaining why 101 different municipalities is probably too much for my nine county region, but the take-away is that we should have some articulatable concept of decentralization’s lower bound because additional institutional disaggregation isn’t always and everywhere a universally positive thing.

Nodes, Edges and Inter-jurisdictional Coordination

The systems which facilitate coordination between political units in a radically decentralized political order is a second area in need of development. We often talk about the merits of places like Singapore, Hong Kong, Shenzhen, etc, but rarely discuss how inter-jurisdictional coordination in a world where small scale polities predominate would actually work.

If we think of polities as nodes in a network, we should consider the lines that connect them as well

Benjamin Barber’s Global Parliament of Mayors and the U.S. based National League of Cities represent an attempt at the type of coordination I’m referencing, but it’s not clear that such organizations could coordinate things like major transportation infrastructure or defense. By and large, these are areas still dominated by states and we may not get a chance to experiment with alternative arrangements until necessity gives birth to new invention.

Swiss Federalism with its inter cantonal concordats might provide an example. The EU with its notion of subsidiarity might offer lessons as well. Both of these examples, however, are instances where a larger territorial entity exists to coordinate activities of semi-autonomous subunits. Long run, they may have all the problems we already associate with statehood. And maybe that means resorting to confederation or bilateral relationships, but I think that’s still a very open question we should be exploring.

Against Tyranny…and Bad Management

An additional question is how to avoid the gradual centralization of power over time. The easy answer is to invoke Exit as the final constraint against power. But I’m unfamiliar with any account that deals with possibility of jurisdictions agglomerating, thereby blunting the usefulness of Exit as they regain the proportions of a nation state. This begs a discussion of what meta institutions we would need to constantly push individual jurisdictions toward that Goldilocks point between prohibitively high transaction costs and diseconomies of scale.

The answer to this question might directly relate the larger question of how individual jurisdictions interact. Or it might have to do with features hardcoded into the individual jurisdictions themselves (perhaps something like Tom W. Bell’s Corrective Democracy). But this is another open area in need of exploration. And I sincerely hope we spend time blazing new trails as we advance our understanding and continue advocating for reform.