Calculation Problems

Ludwig von Mises pointed out as early as 1920 that rational economic calculations are impossible under a command economy. With prices set centrally at will, no rational decision making can be made. Without a free market for goods and services, scarcities cannot be accurately detected and corrected. No amount of surveys and research can take the place of a free market.

This is due to the fact that the market consists of every single person in a society, while a central bureaucracy necessarily must be but a small fraction of the total market. A small set of individuals cannot accurately determine the needs and preferences of all people.

One way of looking at this is that every person in a free society acts like a neuron in a network that encompasses the entire society. Each person transmits signals as to his or her needs and desires so that the overall calculation can be made trough a free exchange of goods and services. Any attempt at a simulation of such a network based on substantially fewer neurons will necessarily fail.

The problem is not merely one of calculation. It is also one of detection. Even if equipped with massively advanced supercomputers, the bureaucrats would be helpless in their search for correct prices because they would be lacking in inputs. There's no way of sampling all needs and desires. We may desire one thing one moment, and something else a little later. These are real time preferences that cannot be registered and fed into a computer.

The free market is in essence a computer, and no computer made by humans can ever come close to what society is by default. By extension, no amount of data gathering by government agents can ever predict the future actions of their subjects. There is little to fear from organizations such as NSA and CIA. They can only monitor a small number of people, and the ones they monitor closely are still impossible to predict. We are therefore free to do whatever we please, as long as some basic precautions are taken. The idea that government spies can find out what we're up to is just as much a myth today as it always was. They may have a lot of data on us, but they have no way of making sense of it all. It's simply too much and too complex for anyone to figure out, even when in the possession of the latest and greatest in computer technology.

Taking this line of thought one step further, we can view the universe as a huge computer in which every particle is a neuron capable of receiving and emitting signals. From this it's clear that we cannot ever hope to produce a computer that manages to simulate anything but very trivial processes. Long term weather forecasts will for ever be uncertain because the number of particles involved are trillions of times more numerous than the computing transistors used to simulate them.

This reduces our scope of predictive power to general trends and trivial certainties. We can be quite certain about things taking place in our immediate future. The rain outside will probably continue for a few more hours. What happens after that is less certain. On the long end we can also be relatively certain that our local climate will stay more or less as it is. What the weather will be on this day in Porto a year from now is on the other hand anyone's guess. It may be dry or rainy. It may be hot or cold. It will not snow, and it will not be scorching hot. That's about all we can say.

The more we limit our scope in time beyond our immediate future, the less certain we can be about what might happen. However, once we allow for broad time horizons, we can say quite a few things with confidence. Political instability appears likely. Economic uncertainty is also a pretty safe bet. We can therefore assume that the price of gold will continue its upwards trajectory. The exact shape and form of future events are unclear, but some consequences are close to certain.

Likewise when it comes to astronomy, we can be relatively certain that Jupiter will eject another moon at some point. However, when this is going to happen is far from clear. It may happen this decade, or it may happen thousands of years into the future. Again we see that detailed predictions are impossible, even if the trend and overall mechanisms involved are clear and well understood.

Jupiter ejecting a moon