2020-02-06

Decentralization and the State

There’s been a lot of talk about decentralization recently in west coast semi-political discourse, with Bitcoin, Ethereum, Urbit, the Sovereign Individual, Seasteading, and many other ideas. I almost wonder whether it is being pushed in a centralized way! This is usually concieved of as an attack on the state, but I want to examine an alternate way of looking at this that synthesizes the impulses behind decentralization with the interests of the state.

To avoid unpleasant preamble, I shall start with the thesis: there are many ways in which the state as such is empowered or simplified in positive ways by divesting power out to decentralized fringes, or by divesting the power of internal rivals. And in general many confluences between wise statecraft and decentralization, and dependencies on the state for decentralization to work. Let’s go through some examples:

The state has to maintain political order, which means either nationalizing or dissolving upstart power centers, so that they don’t become obstacles or even rebels. For example, facebook is an upstart political power, because it can basically censor and editorialize how you interact with your friends. It is coming under the spotlight of political factions trying to sieze that power, and of the state who wants to make sure there aren’t big uncoordinated power centers running around.

One of the best ways to tackle this kind of problem is to force some kind of decentralization on the entity in question. A bit of top down pressure and threat, a bit of bottom up demand for decentralization, and the state is able to eliminate a potential rival. Decentralization of facebook could look like Urbit. If the Urbit guys are wise on strategy, this should be one of their long-term plays: unlike facebook, they can’t become a problematic power center, and so the state could ask the question “why do we need you, zuck, when there are more decentralized approaches to social networking that are more in line with the American way?”. Of course for this to work, the decentralized alternative has to actually be available. We’re all waiting, Tlon.

Besides the stabilization of political order, which is good, decentralization of these rivals will often be a good thing in other ways. Decentralization often means codification into an open protocol that has more staying power. We saw this with trains, oil, electricity, and telephones. Those industries aren’t perfect, but they are fairly strong and permanent institutions that aren’t about to become abusive monopolies. Further, be divesting power from the monopoly, the users usually gain a bunch of rights, like due process and so on, that come from the thing being a “public” good, subject to the discipline of the state. The state, gangster as it may be, is usually much better developed in its restraint than upstart and especially entrenched non-state monopolists.

Sometimes, these large internal quasi-monopolistic entities aren’t a problem politically or even economically. Maybe they’ve reached a stable deal with the state that is both economically efficient and politically stable. Land is a great example of this kind of thing. Large landlords and plantation owners are in a possession of an essentially monopoly rentier good. They do a bit of work, and the capacity inherent in the land comes forth. Mostly they are in the business of being the guy who gets to collect the rent. Since land is so old, and perhaps for other reasons, this often reaches a stable relationship with the state. Wealthy landowners become powerful members of the state, and basically don’t rock the boat too much.

But there are still problems. Specifically, land is a zero sum good. For every wealthy landowner, there are many disenfranchised workers and renters. People come up with various reasons that this is a problem, mostly abstract reasons like fairness, social justice, etc. But one of the biggest and most important reasons this is a problem, from the point of view of the state, is that these people are a large ready-made client demographic for upstart rebel movements, especially communists. They have nothing to lose, they have any easy target to blame for any problems they have, and they are in fact being exploited.

So, preemptively re-distributing productive monopoly goods like land to achieve a more equitable and decentralized distribution might be a good idea. Those formerly disenfranchised workers and renters now reap the benefits of their own labor, and have no one but themselves to blame for their troubles. It is much harder for rabble rousers to turn them against the established order. The economics of agriculture and housing usually aren’t much impacted by distribution, unlike say Amazon’s logistics, which is inherently a central planning exercise. Also, the new petty freeholder class, able to stash a bit of their own surplus, are then able to invest in growth of various kinds. They might not be any better at this than the big guys, or they might blow the money on TVs and drugs, but sometimes the economic calculus works out positive.

Of course this cuts against property rights, but that’s a cost to be weighed in the more serious calculus of political order. In any case, there are many ways to accomplish such things, for example progressive land taxes, that are much softer and more responsive to local conditions than outright forced redistribution.

Bitcoin is another example worth examining. The jury is still out on how it all plays, but the Nakamoto institute guys have done a good job articulating the maximalist case. One bit that’s been missing is how it might positively synergize with the interests of the state as such:

The state wants power, but it also wants to stay unitary. If the state has one will, then it is much more powerful than if it has more total power but more internal coordination problems. As such, it may be a good idea occasionally to divest power out to a decentralized fringe, where it can’t become an object of contention. The state wants to stay laser focused on its core business of power monopolization. Many things can become a distraction.

One such distraction, in an imaginable Bitcoin statism perspective, is the power to print money and distribute it to crony bankers to “stimulate” the economy. This is not a necessary power. Many states have done just fine without it. It is not a terribly useful power either. Nor is it the sort of power that, if left on the table, will be siezed by demogogues and used for nefarious purposes. But it is the sort of power that, given the internal imperfect coordination of the state, the state cannot restrain itself from using. It is also the kind of thing that creates a new entrenched power structure half in the state and half out: the crony bankers around the federal reserve, etc.

In this story, Bitcoin statism comes to the state and says “Bitcoin fixes this”. “We will throw this particular power into mount doom, and you will never have to deal with crony finance again.” If Bitcoin actually does deliver on a permanently sound monetary system, and sound money can deliver economic stability and growth, this might be a good idea. It forces an additional bit of fiscal discipline on the state, and divests the power to micromanage economic cycles, but these can arguably be seen as good things. It also offers some new powers of financial surviellance through the public blockchain, and financial imperialism through uncensorable global money flows, of course.

Bitcoin also claims to be inevitable, but I’m sure it would be a lot more inevitable if the state were on its side, and a lot less inevitable if the state decided to squeeze it. I don’t know definitively whether Bitcoin is good for the state, but I suspect it is, and in any case it’s a great thought experiment and case study of decentralization from the perspective of the state. Bitcoin maximalists should be articulating this state perspective more.

The next major examples of decentralization are localism, heterogeneity, internal experimentation, and competition.

Competition is fairly obvious. Many things are natural monopolies, or can easily fall into monopoly dynamics. Monopolies can get lazy. Even if they stay as good as they would on a competitive market, when they do inevitably decay, they take the whole industry down with them. The state has an obvious interest in making sure there is healthy competition. And competition often needs state support to create free markets. Though of course in some cases, the calculus works out the other way, and the state wants to work with a national champion company.

I’ve talked about localism before, where I argued that globalization leads to fragility. You need some level of replication and artificial supply chain shortening to maintain robustness to random problems and disruptions. Otherwise, you can get a late Bronze Age collapse situation. This isn’t the only reason, there are a lot of benefits of localism, but localism doesn’t win on the greedy efficiency-chasing global market. It needs some help from above. The state is the natural entity that is both interested, and in a position, to protect some level of localism.

Heterogeneity is similar, but instead of cascading vertical failure of a complex system, it’s cascading horizontal failure by common failure mode. Agricultural monoculture is the most well known example: if you have fields and fields of genetically identical crops, the slightest pestilence wipes out your whole crop. But if you have diversification into many different genetic lines, different crops, these monocausal failures are impossible. There are analogues in human culture, institutional design, etc, where a common failure mode rapidly becomes exploited or fails in the same way. The state historically has been the driver of these perverse standardizations, especially in the industrial period. Power is typically abused before we learn how to use it well. Once the power to homogenize exists, the only thing holding it from being abused is the discipline of the state. Fortunately, again, the state has a strong interest in managing these things well. Assuming it can muster the wisdom.

A similar case again can be made for local experimentation. The utility to the overall state of having experiments exploring different possibilities is obvious. But without the state, there will often be no niche in which the experiments can develop, before they become competetive with existing practices. Research and experimentation often depend on explicit protection and sponsorship. This is another reason for the state to sponsor decentralization, and another reason it can’t really happen without that sponsorship.

Many people come up with schemes whereby the state’s power to centralize will be limited by new defensive technology that allows social or economic localities to isolate themselves from outside dependency and interference. Cryptography, bitcoin, 3d printing, guns, nukes, etc. It would take a whole article to refute this thoroughly, but roughly speaking, consider the following:

I conclude from these kinds of thoughts that decentralization cannot be accomplished against the state. Rather, can only be accomplished with the state, and that the state will often be a big sponsor of productive decentralization!