Freedomain Radio

Game theory is often cited to criticize free market economics and demonstrate the need for government regulation.  I know that game theory is in no way a valid argument for government, but I'm having trouble articulating the reasons why. Can anyone help? Or does anyone know of some quick reasons why such an assumption is incorrect? Thanks.

game theory, from my understanding(quick reading on wikipedia), is about making a model that can be followed to forcast future reactions to peoples situations.

if falls apart in economics and social sciences because peoples preferences change every minute (even though it is likely a small change) and over the course of years will be no where near what the model predicted

here is a great example:

http://globaleconomicanalysis.blogspot.com/2008/02/us-homeowners-confound-predictions.html

basically banks modeled (game theory? correct if this does not seem correct) that people would default on credit cards/car loans BEFORE home loans. but peoples ideas change and the model falls apart.  any kind of economic model WILL fail when peoples ideas/opinions change, and they always do.
 

I'm not an expert on game theory, but here's my response, based on what I know about it.  If someone could correct me, I'd greatly appreciate it  

The first key is to understand the assumption involved in game theory.  I'll use the "prisoner's dilemma" as an example.  In the prisoner's dilemma, Larry and Barry are being questioned in separate rooms as to their involvement in a crime.  They are both guilty of the crime, though the police do not know this.  The decision Larry and Barry must make is whether or not to rat out the other.  If Larry rats out Barry, then Barry goes to jail for 10 years and Larry goes for 5 (in exchange for his cooperation), and vice versa if Barry rats out Larry.  If neither rats, they both go free; if both rat, they both go to jail for 10 years.  Clearly, the best outcome for the two is that neither rats, and they both go free.  But, they are in separate rooms, so they do not know what the other player will do.  If only there were someone who could make the decision for both Larry and Barry, the "optimal" outcome could occur every time.

So it is with the economy, according to this game-theory critique of the free market.  There is limited knowledge available to the many parties in an economic system, and if it were possible to increase knowledge -- or, better yet, invest in one group the authority to plan the economy altogether --  then better, more rational, more efficient, more productive results would follow. 

Take producing a pharmaceutical.  The interests of the consumer and the producer are, say game theorists, at odds: the consumer wants a safe, effective, inexpensive drug, and the producer wants to make profit.  One way to increase per-item profit is to forgo safety procedures in the manufacturing process, to leave impurities in the drug, or to use a less potent version of the active ingredient.  The game-theory regulation can take the form of FDA regulations to let the consumer know that the drugs they take are safe and standardized.

The major unstated premise is that the government can perform this task better than the free market.  If you attack that point, I think the argument from effect put forward by game theory falls apart.

The main problems with game theory are that it is inaplicable to the real world.  First, it has the problems of utilitarianism, in that interpersonal utility cannot be objectively compared and that it is not a moral argument.  Moral, not practical arguments are what convince people.  Secont, the situations handled by game theory are to simplified to be applicable.  Third, the statist interpretations of game theory ignore the complications of applying the same arguments to the state (see public choice theory).

It's just a tool, man. You can use the tool correctly to invalidate the state or use it incorrectly to validate it.

Logic is used by some to validate the state. One should not be concerned with tearing down the entire construct of logic, but pointing out the flaws in  individual arguments.

If you have any specific game theory arguments you would like to address, I would be happy to help you tear them up.
 

Nathan Miller:

Third, the statist interpretations of game theory ignore the complications of applying the same arguments to the state (see public choice theory).

It is somewhat telling that the foremost example of game theory application, the prisoner's dilemma, is totally immersed in a statist setting, the possible effects of which are not at all mentioned.