Autarky is unstable

Autarky means economic self-sufficiency. In the past many countries tried to depend as little as possible on trade with the outside world. This can be done, to some extent, by a country, especially a large country with extensive resources. Some people who believe in the nanotechnology meme claim that it will soon be possible for individuals to be self-sufficient. According to this scenario, everyone will have a desktop manufacturing machine, also known as a "box," which will produce everything he or she needs. In this scenario, there is no trade.

The question to ask is: why is there trade now?

In this section I am going to assume, for the sake of argument, that it is possible to compress the entire economy into a small self-contained system. Even with this assumption, the Autarky scenario still will not happen, and cannot happen. Autarky is inherently unstable.

Suppose we have an economy in which all individuals have home nanotech facilities. So our starting point is an economy similar to early feudalism, when it first emerged from the ruins of the Roman Empire. Each local Baron had his own farm, his own orchard, his own cattle, etc. He and his village built their own houses and made their own clothes, their own horseshoes, everything they needed. They even made their own tools. A Baronial estate was a very coarse-grained version of a genie machine. It could produce anything, if "anything" is construed in a generic sense: some kind of food (not necessarily very good), some kind of clothes (not necessarily very good), and so forth.

Our MNT-Autarky scenario is a high-tech version of this, in which each individual has a genie machine based on nanotechnology, which can produce anything the world economy can produce. There is no need for trade.

The Baronial estates didn't remain self-sufficient. They traded among themselves, and over a period of many generations trade became more and more extensive, until the feudal economy gave way to the modern global economy. I think it is a theorem in economics that this must happen.

Before stating the theorem, we need a definition. I am going to assume that the home nanotech facilities are finite. By "finite" I mean:
1. The home nanotech facilities use energy (they are not perpetual motion machines).
2. The home nanotech facilities perform their tasks at some measurable rate (not instantaneously), and different tasks take different times.
3. The cost of retooling or reprogramming a machine is nonzero, and the cost varies depending on what kind of retooling is done.
4. The home nanotech facilities have an interface that requires some expertise and effort on the part of the user. Different tasks requite different degrees of expertise and effort. Home nanotech facilities have to be programmed, rather than merely commanded -- or, if they can be commanded, the commands have to be so detailed that it is a lot of trouble to give them properly, so "commanding" amounts to the same thing as "programming."

The Gradient Theorem.
If the home nanotech facilities are finite, then the MNT-Autarky scenario is not stable. Any approximation to it is unstable. In fact, there is a gradient away from self-sufficiency, in the direction of more specialization and more trade.

Informal proof.

If the output of a home nanotech facility depends to some extent on the effort put into it by its user/programmer, then all home nanotech facilities will not be equal, any more than all humans are equal (or all Baronial estates). Each of us will have his/her own unique home nanotech facility, just like we all have a computer, we all have a set of tools of some kind, we all have a personal library, and so forth. No two are exactly alike. Even if they were alike to begin with -- even if they were based on the same hardware and had the same software installed -- there would still be differences in how people would use them. We all have different skills. There will always be differences in the effectiveness of different individuals at performing a given task, even when they are using home nanotech facilities, just as there are differences in how people use computers.

Every feature of the home nanotech facility will have a learning curve. People will be power users of some features of their home nanotech facility, and never get around to reading the user's manual for many other features. If they need to use a feature that they don't know much about, they will hire someone who is an expert at using that feature.

Therefore trade will exist, for the same reason it exists now.

The existence of trade means that the various home nanotech facilities and their operators will be in communication with each other, offering various services. This will lead to more specialization, not less. Individuals will tailor their nanotech facilities and their skills to the market. They will find that it is in their economic interest to specialize, rather than to try to do everything. Home nanotech facilities will give way to centralized factories. Self-sufficient nanotech estates will give way to an economy based on specialization and trade, just like the Baronial estates gave way to the modern economy.

In the MNT-Autarky scenario, it is assumed that home nanotech facilities are not finite (i.e. they don't satisfy Finiteness Assumption #4). They don't have an interface. That's what makes them genie machines! They don't have to be programmed, and there is no learning curve. Using the genie machine to make a skyscraper requires no effort or skill. I don't even have to think the problem through and write up detailed specs for what I want. All I have to do is say "Make me a skyscraper," and presto, it's there.

If it requires any more effort than that -- in other words, if there is any measurable difference between the effort required to produce different items -- then the situation is unstable. The inequality of effort will create a gradient leading away from self-sufficiency, toward an economy with trade and ever-increasing specialization.

Now, consider the converse problem. Is centralized manufacturing unstable? Is there a gradient leading away from centralized manufacturing, in the direction of do-it-yourself home workshops?

Suppose industrial nanotechnology exists. The year is 2012. Intel and several other companies have chip factories that make nanoscale circuits by putting atoms in place one by one, guided by lasers. What incentive would they have to scale these factories down and make smaller versions of them available to individuals? Can anyone construct a sequence of economically motivated steps leading from industrial nanotechnology to home nanotech facilities?

back to the table of contents

home