Decision Theory in Political Decisions

One would think that with better computational aids, the arbitrary decisions we make via politics would slowly yield to rational analysis. I see no indication that this is happening.

To select one example from many possible, think of the ability of government to manipulate the distribution (and creation) of wealth by tax policy. Assume the goal is to raise enough money to run a slim government and to establish conditions such that a desired distribution of wealth comes about through normal business. That is a lot to assume, but we are not even started.

Is there an optimum distribution of wealth and could it be obtained? That is easy. Start with a forced re-distribution of wealth equally to all citizens. No exceptions. Most people would agree that would be disastrous. At the other extreme, have the government confiscate all money and give it to a single person or small group of persons to do with it as they wish. That would not be good either. Both of these distributions could be obtained by a government in principle, and both are worse for society than what we currently have. With a few more reasonable assumptions, we can apply Rolle’s theorem of extremes to conclude that at least one optimum distribution exists and it can be obtained.

Decision Theory in Political DecisionsMore important, what is the definition of an optimum distribution? Is it like pornography: we cannot define it, but we know it when we see it? The difficulty is due to the non-linear response of the accumulation of wealth. A society with only enough wealth to continue eating is unlikely to spend resources on research and development to improve matters and incidentally create more wealth. A society in which most of the wealth is locked up in a Rajah’s house and used for the creature comforts of a few is unlikely to improve. But in either case, diverting small fractions of the total wealth toward education and research can have extremely beneficial results.

Any fiscal conservative at this point might object strenuously and say government has no business attempting to change the natural distribution that comes about from free markets. An extreme socialist would have the opposite argument that instead of imposing taxes with the hope of directing resources into desired paths, the government should simply take over the means of production and make them function in the desired way.

We have many examples of why neither approach works optimally.

Assuming people of good will get together (by far the most shaky assumption I have made) and agreed on the ideal goals of a society, then in principle, one could derive the function that describes an ideal distribution of wealth. We could answer the question of how many millionaires are ideal and how many billionaires are ideal. There would always be some poor people (better men than I have said that), but the definition of poor would surely evolve. The tax code could be written such that on the average (!) that distribution is obtained. No coercion needed. Businesses and individuals would behave in their own best interests as they do now, but the results, on the average, would be better for everyone.

This is obviously an exercise in how to set up a decision process. Please do not assume I am making a political statement. The only political statement I make here is that nothing this logical will happen. I also predict that no politician will ever stand up and state what an optimum distribution is. Democrats will bemoan the loss of the middle class and accumulation of wealth in a few. Republicans will talk about free markets and job creators and how we all benefit from the work of rich people. What should we make of these arguments? Shakespeare said it best: “It is a tale told by an idiot, full of sound and fury, signifying nothing.”

Article Written by

  • Stlhobbit

    Your ending quote could be pluralized to be an apt description of our 535 overpaid & over-bribed/lobbied elected representatives in Washington:  “It is a tale (fiction/lie) told by idiots, full of sound and fury, signifying nothing.”

  • bern

    The problem with the discussion above is that there is no commonly accepted definition of “optimum distribution”. The distribution that makes most people “happy” is not the definition that leads to the lowest infant mortality rate, is not the definition that leads to the greatest increase in total wealth, is not the distribution that leads to any other “good” you may define. Rather than letting some corrupt or ignorant bureaucrat decide what “optimum” is, I prefer to let the “market”, that is, a great mass of people, each acting independently as they freely wish determine what the distribution will be through their behavior.