Have you ever wondered why a few people end up owning and controlling the vast majority of a society’s wealth, or why aristocracies inevitably develop in practically every culture, past or present? Have you pondered the egregious wealth inequality in the US where the top 1% owns the wealth equivalent of the bottom 95% combined? Some years ago, I set out to investigate how capitalism really works using a totally clean slate and using new computer programming methods. Over several months during the spring of 2003, I constructed a model economy using statistically equal entities, and I achieved results far beyond my initial expectations. It turns out that huge levels of wealth inequality are built into capitalism-even among statistically equal people. Capitalism takes the wealth generated by the society and then re-distributes it to a tiny minority.
My model economy programs use stochastic (i.e., equal chance) methods with economic agents who own equal amounts of wealth initially. Economic agents follow the following simple paradigm:
Agents make investments.
Agents make monetary exchanges.
Agents receive returns on investment.
Using just the three points above, a decent approximation can be made of what an actual capitalist wealth distribution will actually look like. Through trial and error, and by reconstructing the process that would tend to maximize individual agent relative wealth, I found that an investment hierarchy in addition to the above points generates correct wealth distributions. The “investment hierarchy” that I am speaking of represents the employer to employee relationship (especially corporate management): in the model economy as in real capitalism, employer investment money becomes employee labor income. In turn, the next lower tier employee becomes the employer and source of labor income to the tier of employees below him, and so on and so forth. After the hierarchy makes their investments, monetary exchanges take place for goods and services, and then return on investments filters back up through the hierarchy. This methodology produces astoundingly accurate results. Not only can one accurately predict personal wealth using this algorithm, but it also produces correct stock market capitalization (i.e., corporate wealth) distributions. The power of using this method comes from its predictive power. For example, one may now answer the following question: What will be the long term distribution of wealth of any large group of people participating in capitalism without taking taxes into consideration?
Answer: According to the model,
The top 1% will own 35% of the wealth
The top 5% will own 64% of the wealth
The top 10% will own 75% of the wealth
The top 20% will own 84% of the wealth
The bottom 60% will own 5% of the wealth
The bottom 40% will own .013% of the wealth
If you keep track of such things, then you may have noticed that the above computer generated predictions match the current US wealth distribution almost exactly. However, I also said that the predictions don’t take taxes into consideration. It turns out that taxes in the US have no re-distributive effect whatsoever. Our tax system guarantees that wealth will be as concentrated as possible, and that after taxes that wealth will still be as concentrated as possible.
Let me make a few more observations about capitalism as illuminated by the model economy:
The investment process itself causes wealth inequality for the entire population. Moshe Levy of the Jerusalem School of Business proved that success in investing was independent of investment ability for the wealthiest people. My paradigm generalizes this idea for the whole economy because it takes trade into account, and because it shows that all people are investors.
If people are equal in every respect, and if they can make even the most basic decisions to increase their wealth, then a wealth distribution such as the one I have described has to develop. This distribution appears in every real free market capitalist economy
All members of an economy are investors. Even though people at the bottom get most of their money through labor income, they still have to invest at the same rate as the wealthiest people if they are going to maintain their relative wealth (i.e., their piece of the pie). An example of how an ordinary person invests could be: a person pays someone to remodel their kitchen, and then their house is sold for a profit or loss when taking the original house cost plus the remodeling cost into account.
The neo liberal idea that if the trade rate increases, then the wealth inequality decreases is true, but it doesn't work well in a real economy. If there were no hierarchy (bosses and workers), then neo liberalism would work (but the environment would be damaged since increasing the investment rate increases consumption proportionally). The hierarchy's effect on increasing wealth inequality far outweighs trade's ability to decrease wealth inequality.
The vast majority of wealth inequality is caused by the hierarchy (bosses and workers). However, it would be impractical to remove the intermediary agents between investors and producers because economies have large numbers of people, and single people would have great difficulty making exchanges with tens of thousands of people.
Progressive taxation is the best way to reduce wealth inequality in a free market capitalist system. The easiest way to do this is to tax the income of the wealthy, but a wealth tax would work too. Taxation though, has to redistribute wealth, or it won't work.
There is much more to all of this, but I have run out of time tonight to explain it all. Thanks for taking the time to read this article.
I am a Social Democrat and a former congressional candidate who lives in Columbia, SC. I'm a fairly accomplished programmer (in LabView) who has done much work with statistical distributions in the electronics and fiber optics industries.
Residual income from existing assets probably cements the percentages you cite.
If someone suddenly gave you 2 million dollars, and you simply supplemented your regular job with the interest you would annually make off of your 2 million dollars, you might never be poor again if you didn't live an austentatious life.
by
Alessandro Machi (13 articles, 0 quicklinks, 1 diaries, 174 comments)
on Friday, July 13, 2007 at 8:07:36 AM
I enjoyed your article on capitalism. If we were not all so consumed, government endorsing consumption mind you, with the financial obligations of capitalism (in which the working class ultimately toil for the wealthy in an effort to join their ranks or to be pacified by them) perhaps the inequality seemingly built into it could be addressed.
by
Andrea Lewis (0 articles, 0 quicklinks, 0 diaries, 1 comments)
on Friday, July 13, 2007 at 8:08:13 AM
Would very much like to see the model and how it works
I read somewhere something once said that achieving a millon dollars in assets is next to impossible, however, once done, making two million dollars (absent stupidity) is virtually assured. That increases geometrically as wealth increases. In other words, if you make 10 million dollars, making 25 million is virtually assured. If you make 25 million, 100 milliion is virtually assured. Etc.
The impossible dream is getting to the point where you have enough capital to start the engine, so to speak. You "cannot get there from here" as any sort of "regular" wage earner.
by
Steven Leser (212 articles, 45 quicklinks, 33 diaries, 1391 comments)
on Friday, July 13, 2007 at 3:25:53 PM
The New Scientist article "Why it is hard to share the wealth" in conjunction with "Now that's power" makes for aninteresting juxtaposition. The implication is that the richest 0.1% ofthe population's income is described by Pareto's Law -- meaning that ifyou're born into that kind of money, you'll only get richer, no matterwhat you do. And the poorest 99.9% of the population's income isexplained by Boltzman's Law -- describing random movement of gases in anenclosed area, and meaning that even if you've struggled up to the leftend of the Pareto curve, you're far more likely to get poorer than tomake it to the 0.1% elite. So much for the American Dream.
by
Bleeding Heart Liberal (0 articles, 1 quicklinks, 0 diaries, 48 comments)
on Friday, July 13, 2007 at 8:21:57 PM
Everyone: many thanks for your comments. I would love to give you a copy of the model economy program, but unfortunately the programming language that it is written in is proprietary and it has high licensing fees. However, I surely will convert the model into another free programming language sometime in the next few months. I would have done this long ago, but I’ve been so busy at work that I just haven’t had the time to do much of anything else.
Yes, I do assert that wealth tends to remain static and unequally distributed-and that’s a good starting point for a more technical description of how the model works. Perhaps the best way to explain this is by reviewing the reasoning, step by step, that I initially used to create the model.
Back in 2003, the first thing I did was to create an array of one thousand agents, and I gave them each ten dollars. The program would select the first agent in the array and then select another agent at random in the array. Each agent would take the same percentage of his wealth and then exchange with the other. The program would click through each agent in the array and perform the same process. After the last agent in the array performed an exchange, then the entire array was sorted in order of wealth. The process repeats itself over and over again, and a distribution soon developed (also the same distribution developed each time, every time). The program took the distribution over time, and averaged it for each iteration to give a solid, repeatable distribution of wealth. At the time, I didn’t even know the name of the distribution, but after some research, I found that the distribution is called a Boltzmann-Gibs distribution. This is the same distribution that appears concerning particle collisions that is used in physics. It’s an exponential distribution, but it is nowhere nearly as concentrated as the real distribution of wealth in capitalism.
After developing this initial distribution based only on monetary exchanges of equal percentages of individual agent wealth, I started trying other scenarios. What if the percent exchanged was selected at random, or what if an agent would only make the most advantageous exchange from a number of possible exchanges? After trying many different scenarios concerning exchanges, I found an important truth: the Boltzmann Gibs distribution is robust. It doesn’t matter how much or in what percentage that people make monetary exchanges because the B.G. distribution will always develop over time. The important thing is that all people make exchanges, and then the B.G. distribution will necessarily develop.
Equally important though, is the fact that although the same exponential distribution always develops, the people filling slots in the distribution continually change. Using just monetary exchanges, wealth continuously shifts from person to person. One moment you may be rich, but in a very short amount time you may be poor in a never ending cycle.
This fact gives me insight into how capitalism must have developed. Long ago people must have figured out this boom-bust cycle concerning their personal wealth. Once people became rich, they probably started looking for ways to preserve their wealth because they knew from experience that if they kept making monetary exchanges without an advantage, that they would soon become poor again.
Consequently, I programmed an advantage in the system in the form of capital investment. So now in the program each agent in the array would give a portion of his wealth to other agents and receive back proportionally the gain or loss from the agents that he gave money to. Every agent in the array still makes monetary exchanges.
An amazing thing happens when capital investment is added onto the monetary exchange system: wealth becomes much more concentrated and people tend to become more static within the distribution. Rich people tend to stay richer and poor people tend to remain poor.
For a long time I tried to find how much a person would have to invest, and how many people he would have to split his money among to maximize his own wealth. I found that the agents would have to invest half of their wealth between two other entities to maximize their return in the long run. Based on this information, I surmised that income should be about half of wealth-a fact that I latter confirmed through the Federal Reserve Triennial Survey. Since the survey began, I found that median wealth is almost exactly twice that of median income (1.97:1).
So, with exchanges and capital investment, it is possible to construct more realistic wealth distributions, but there is one more piece to the puzzle. Up to this point, the distributions are lopsided, but not lopsided enough to match real capitalism. The distribution at this point is almost the right shape. Thinking about it though, how would one even further reduce his risk in conjunction with capital investment? If I’m on the top and if I want to reduce my risk, then I just don’t invest in anyone, I invest in people who are just below me in relative wealth, and they in turn invest in people just below them in relative wealth. Regardless if people below me gain or lose money in the aggregate, I always gain the majority of what was gained.
That’s what makes capitalism so attractive to the wealthy-they’re pigeonholed to win because the hierarchy limits their risk-the wealthier you are, the less risk you are at. You’re bound to remain on top. That’s why you or I will never switch positions with Bill Gates or the Wal-Mart people.
So in the program when a hierarchy is added to capital investment and monetary exchanges, then the correct distribution develops. It’s only beneficial for people down to average wealth though to continue participating in the hierarchy because after that, no monetary gain is realized for the agents. Average wealth in a capitalist system starts at the top 20%, so in the US that means people with four year degrees and above who a part of a corporate hierarchy benefit from the system, and through history aristocracies and what was latter called a middle class (called upper middle class today) have always involved roughly the top 20% of their respective societies. Capitalism, regardless of what textbooks tell you, has been around for a long time even though historians have called it aristocracy.
As a final note, in addition to the wealth percentages that I cited in the original article, I found the natural Pareto coefficient to be 1.76. Also, although I don’t have the time to get into it, in 2005 I found the real name of the wealth distribution. It’s the Woods Saxon potential curve (which looks a lot like a Bell curve). The log of wealth vs. the number of agents ranked from richest to poorest is in the shape of a bisected Woods Saxon potential curve. Incidentally, I also found a shell structure concerning wealth that closely mimics nucleon shell structure that is used in nuclear physics. Thanks for reading this.
Sincerely,
Mark Whittington
by
Mark Whittington (1 articles, 0 quicklinks, 1 diaries, 22 comments)
on Saturday, July 14, 2007 at 1:51:42 AM
The Tsallis distribution might be worth looking at.
If you are looking for an open-source modeling language I recommend Python - there are faster, more efficient languages, but Python is easy to develop in and it has some very nice tools (such as scilab).
by
cam (0 articles, 0 quicklinks, 0 diaries, 54 comments)
on Monday, July 16, 2007 at 3:35:11 PM
Must Read (
Well Said (
News (
Touching (
Funny (
Supported (
Interesting (
Inspiring (
Valuable (
