Sunday, November 4, 2012

Summary of "Self-replicating economics"

Last week at work, I unfortunately had to sit through a presentation  by an economist outside our company who tried to give us a 'Crash Course in Economics for Engineers.'   Whoah!

Have you ever watched or sat through a presentation by an economist and had the thought jump into your head, "This person has just completely talked themselves into a circle. They started out the sentence explaining why inflation is good, and by the end of the sentence, they finished with how deflation raises the value of the dollar and, hence, deflation is good."  Note: it scares me when economists ignore the self-referential nature of economics calculations. In general, if you try to ignore self-reference, it will show up and make you look like a fool. (For example, thinking that government spending has a multiplier effect greater than one.) It's better to just include the self-reference into the calculation from the start. Physicists learned this years ago when they asked the question: what is the charge on a bare-naked electron if there were no vacuum polarization? Chemists had to ask this question when they were studying the collective bonding effects in polar liquids like water. In both cases, scientists resorted to renormalization theory to remove the infinities that would pop up. Every economist either needs to understand renormalization theory or needs to take a few courses in biology, i.e. the study of self-replication.

Here's what I supposedly "learned" in that presentation:
(1) The business cycle is cyclic, and the next peak will be in 2015.    (Don't worry. The business cycle is not predictable and is not sinusoidal.  Many people have shown this, but if you want proof that there is no such thing as a business cycle, check out my Frequency Analysis graph in one of my earlier posts. There is no one frequency that stands out above the rest, except the junk at low frequency due to the overall, non-cyclic trend of increasing GDP. So, don't believe anybody who predicts for you when the next GDP peak will be, based off of when the last peak was and based off of some seven/eight year cycle.)
(2) Economies fluctuate about a steady-state growth rate. Steady-state growth is the equilibrium point and oscillations about this point are nearly perfectly sinusoidal.   (Don't worry. This is false as well. There is no such thing as the 'steady-state growth rate.' And as mentioned above, oscillations in the real world occur at every conceivable frequency.)
(3) The economy is a closed system because the money just keeps on circulating, so there's no need for a Federal Reserve to keep printing money.   (There's at least one problem with these statement. First, the economy is not a closed cycle. It's like an ever expanding spiral in which useful work is used to make more useful work. Second, the amount of money in circulation should ideally increase as the spiral gets bigger and bigger, i.e. as we make more and more power plants. Money is simply a way of determining who gets to use what percent of the useful work being generated from power plants. Theoretically, it shouldn't matter how much money is in circulation, but in practice, most people prefer that the prices of items they buy remain constant or decrease with time, while at the same time, they prefer that the dollars in their bank accounts increase with time. Ideally, there would be a non-profit agency within the government that forces the Treasury to print or remove enough currency from circulation in order to maintain 0% to 2% inflation in the average price of useful work. Right now, that agency is Federal Reserve. But it isn't really a part of the government and it hasn't been doing a good job over the last six years because there are some confused economists on the Board of Governors who think that the Federal Reserve can decrease unemployment by printing money and giving that money to the government and Wall Street. While the Federal Reserve system in the US has some major flaws, these flaws could largely be fixed by adopting a rule-based system of when to print or remove money from circulation, as has done by many countries, including Germany before the euro.)

Given the fact that economics can be an extremely difficult to teach, the goal of this post is to summary the main points I've made over the last few years in the area of "self-replicating economics," i.e. the study of human-designed system that can self-replicate nearly-independent of biological systems. My hope is that economics can become a fairly easy subject to teach if instructors focus on examples of small, self-replicating systems.

For example, we have designed "oil/gas well, power plant, steel mill" systems that can self-replicate nearly independently of other systems. In other words, we can drill wells, build steel mills, build factories, build vehicles, build turbines, and build power plants in ways such that the whole system can self-replicate and grow. Unfortunately, the growth rate of this "oil/gas well, power plant, steel mill system" has been decreasing slowly over time (despite increases in technological know-how of the people designing the systems) because the increase in technical know-how has not off-set the increase in direct and indirect taxes on this system as well as the fact that the oil/gas we drill for is often at lower and lower depths and that the systems we design now are often more complicated in order to reduce the damage on biological self-replicating systems.

I think that it's important to focus on and teach all of the main components of a system that can self-replicate independently of other systems. In other words, if you are going to teach the subject of economics, don't focus just on small parts of the system, such as hamburgers and soda. People, hamburgers and soda don't represent a full system that can replicate. A full system might be:  traders, homes, clothing, police, politicians, cattle, goat, blacksmiths, farmers, civil servants, teachers, chefs, horses, etc... Since the smallest set of items in a self-replicating system that includes human is probably really large, it might make sense to start with simple mechanical systems that are complete and self-replicating. For example, in prior posts, I've discussed possible examples of human-designed self-replicating systems, such as a self-replicating wind turbine/factory systems and auxons, which are self-replicating solar panel/factory systems. It's easy to talk oneself into a circle if you don't study the entire self-replicating system. For something as large as the world economy, I understand that it is difficult to see what's really going on (i.e. self-replication), so what I normally try to do when teaching economics is to simplify the economy down into its main components so that one can see that economic growth is due to the fact that he have design "oil/gas well, power plant, steel mill, turbine factory, electronic equipment factory" systems that can self-replicate. Before we designed these thermo-mechanical systems, economic growth rates were pretty much confined to population growth rates (of plants and animals in the system.) The goal of the remainder of this post is to summarize what I call the field of "self-replicating economics," which is the study of complete, independent systems that can self-replicate and grow.

(1) The purpose of a "power plant / factory" system is to self-replicate as fast as possible without damaging other self-replicating systems. The power plant is just one component of the self-replicating "power plant / factory" system. Whenever I stated previously that the purpose of a power plant is to make more power plants, what I meant by "power plant" was the entire "power plant / factory" system.

(2) For any individual component in the system (such as the turbine factory or the steel mill or the gas well or the natural gas combined cycle power plant), the purpose of the individual component is to achieve a large, positive rate of return on investment by making and selling products into a free, unsubsidized market. Since the product from the power plant is electricity and has units of useful work [kWh], we can say that the purpose of the power plant is to obtain a large rate of return on useful work invested. A power plant should generate much more useful work (i.e. electrical or mechanical work) than the useful work consumed to build, maintain and fuel the power plant. The power plant in the power plant system is like the mitochondria of a biological cell.

(3) A wise investor is an investor who owns a little bit of (or all of) each of the components of a complete system that is capable of self-replicating. This is what we mean by decreasing portfolio risk. If you own a natural gas power plant, you run the risk of an increase in the price of natural gas. But if you own the gas well, the turbine factory, the steel mill and power plant, then you own all of the components and all you need to do to increase your wealth is to make sure that each of the components is communicating with the other parts of the system. If you can't afford to own an entire self-replicating system, then perhaps you could invest in 5% of a power plant, 5% of a gas well, 5% of a turbine factory, 5% of a steel mill, and 5% of a electronic equipment factory that supplies the parts for the power plant. A wise investor doesn't invest in just one of the pieces of the self-replicating system. On the other hand, the wise invest should be smart enough to know what is essential for the system to replicate (such as the actual power plant and the factories that can build the components of the power plant) as opposed to what is non-essential for the system to replicate (such as gum, Facebook, Cabbage Patch Kids, yachts, hotels in Greece, exercise bikes, and Monet paintings.) Don't get me wrong. If you are wealthy because you already know how to invest in self-replicating system, go ahead and purchase these luxuries. Just make sure that you know the difference between systems that can self-replicate and luxuries. Luxuries can't self-replicate; they are items that suggest to other people that you know how to invest in projects that can self-replicate. However, people can get into trouble when they mistake luxuries for systems that can actually generate positive rates of return on useful work invested.

(4) People, animals and all life forms are self-replicating power plants. In fact, humans and other animals are self-replicating fuel cell systems. Plants are self-replicating solar cells. The purpose of a biological system is, like any thermo-mechanical power plant/factory system, is to obtain a large rate of return on useful work invested, i.e. to self-replicate as fast as possible without damaging other self-replicating systems.

(5)  A human can exist without ears or without toes, but a human can't live without mitochondria.  The mitochondria are the power producers of the body. Mitochondria, like the earliest lifeforms, have a large positive rate of return on work invested. In other words, the amount of useful work to build a mitochondria and to look for food is less than the amount of useful work that the mitochondria can generate from the food that the specie finds.

(6) If we presently generate TWs of useful work, and if life forms (and their power plants) have been increasing the amount of useful work as time has gone on, then there was ultimately a point in time in which there was only a small amount of useful work being generated by life forms. So, this begs the question. When was the first generation of useful work? How did life arrive out of non-life? How did the eternal cycle of growth begin? (i.e. how did the cycle of constructing cells, storing useful work, spending useful work to search for food, generating useful work from food, and then reproducing begin?) I'm interested in knowing what sets of differential equations can produce growth of self-replicating structures. (Please email me if you know of any work being done on predicting when a certain set of differential equations yields self-replicating structures. So far the only structures that I have only seen emerge out of differential structures are stationary attractors, N-dimensional limit cycles, and strange attractors.)

(7) We can't correctly predict the future of the economy because there is no complete solution to systems of differential equations that represent systems far-from-equilibrium. All we can do is to make simplifications and hope that our simplifications don't leave out parts of the system that have large effects on the question: can the system self-replicate and grow?

After having read the bullets above and some of the previous posts at this website, you will hopefully appreciate the simplicity of teaching economics when you focus on a complete system that is capable of growing through self-replication. To teach economics, it's important to pick a system that is not too small (i.e. does not have all of the elements required for growth) and not too large (i.e. has components that are non-essential and just add to the complexity of the system being taught.)