The Hungarian-born mathematician was a genius among geniuses. John von Neumann was a colleague of Albert Einstein at Princeton's Institute for Advanced Studies. He made many pioneering contributions to a wide range of fields in math, physics, statistics, and mathematical economic theory. He was a key member of the Manhattan Project that developed the atomic bomb.
Von Neumann's wacky side was legendary - his partying, story-telling, and off-color limericks for every occasion. As one author put it, "Who else could help develop the bomb, but also play practical jokes on Einstein?"
During the mad dash to beat the Nazis to the bomb, von Neumann somehow found time during the war to co-author with economist Oskar Morgenstern the first book on game theory.
Typical of von Neumann's eclectic interests, he wanted to figure out a scientific approach to bluffing in poker. In poker, you don't know what cards the other players have, giving players the opportunity to misrepresent strengths and weaknesses.
In other words, you make decisions not only on incomplete information, but also on information deliberately mis-represented. Von Neumann realized that how best to act on these situations would apply to much decision-making in business, and society in general.
Game theory eventually became useful accepted methodology in fields such as economics, political science, and biology. Today, game theory offers insights into practical pursuits, including business strategy.
As people in real life don't behave according to mathematical theory, game theory is not necessarily a predictive tool. Nevertheless, it can guide decisions, and is a useful analytical device for explaining what happens in many real world situations. In particular, it explains why results among multiple participants often end up less than optimal for an individual, and the combined result inferior to what could have been attained for the group.
In 1950, two researchers at the RAND Corporation, a prominent think tank, produced a paper showing mathematically that final solutions are often less than optimal. A mathematician at Princeton University, Albert Tucker, devised a simple story to illustrate this phenomenon, "the prisoners' dilemma."
Smith and Jones are arrested for a crime for which the evidence is murky and inconclusive. If both deny, they get off with a lesser charger, perhaps no more than a fine. If both confess, they share the rap, and do moderate prison time.
They are held separately. Smith is offered a deal - if he confesses, and implicates, i.e., rats on Jones, while Jones denies, Smith goes free. Jones takes the rap and does heavy jail time. Jones is offered the same deal - if he confesses and rats on Smith, and Smith denies, Jones goes free, and Smith does heavy jail time.
To repeat, if both confess, they do moderate jail time. If both deny, they get off easily because of the murky evidence.
So, isn't it obvious that they should both deny? A light fine would be the optimal solution for both combined. But it's not likely to happen.
Smith considers his options under alternative scenarios:
Suppose Jones denies. If Smith denies, Smith gets off lightly, paying a fine. But if Jones denies and Smith confesses, Smith goes free and Jones takes the entire rap, doing heavy jail time. Any honor among thieves, sentiment, morality, or sense of loyalty apart, if Jones denies, Smith has the incentive to confess. But he has no assurance that Jones will deny.
Smith has to consider the possibility, actually, the likelihood, that Jones will confess.
If Jones confesses while Smith denies, Smith takes the entire rap and serves heavy jail time - worst-case scenario for Smith, while Jones goes free.
If both Jones and Smith confess, they share the rap and do moderate jail time. Again, Smith has the incentive to confess. Better to do moderate jail time than to play the sucker, take the full rap, and do heavy jail time.
Note that whatever Jones might do, confess or deny, Smith has the incentive to confess.
Since Jones is offered the same deal, his reasoning is exactly the same. No matter what Smith does, confess or deny, Jones has the incentive to confess.
Net result: Both confess, and both end up serving moderate jail time even though their optimal solution would have been for both to deny. But because neither knows what the other will do, and both have the incentive to avoid worst case scenario - playing the sucker and doing heavy jail time while the other goes free - a less-than-optimal solution results for both.
This "game" gets ever so more complicated and less likely to result in optimal solution when there are more players involved. And it flies in the face of the proposition that each individual acting in his/her own self-interest automatically furthers the greater good.
But surely, the skeptical reader will reason, this "game" drummed up by some ivory tower dreamer has no basis in reality. Surely, rational, intelligent, honorable people would avoid getting into this dilemma - this trap.
Really? Even when there are billions of dollars in profits at stake? So huge Wall Street financial institutions run by super ambitious, clever people with egos big enough to fill an auditorium are about integrity, honor and putting the economic health of the nation above billions of dollars in profits? And individual traders are willing to subject multimillion-dollar bonuses to the broader objectives of ethics, integrity, and a healthy economy? If you believe that, I have a bridge in Brooklyn about which we should discuss.
Wall Street is about making money - lots of it, period. So what if the scions of Wall Street have to invent fuzzy products and new schemes to make commissions, stretch the law, bribe the congress to eliminate regulations, or pass laws unleashing them from constraints? Let someone else worry about the economy.
Next week: Wall Street, the sub-prime mortgage fiasco, and the prisoners' dilemma.
- John Waelti of Monroe can be reached at jjwaelti1@tds.net. His column appears each Friday in The Monroe Times.
Von Neumann's wacky side was legendary - his partying, story-telling, and off-color limericks for every occasion. As one author put it, "Who else could help develop the bomb, but also play practical jokes on Einstein?"
During the mad dash to beat the Nazis to the bomb, von Neumann somehow found time during the war to co-author with economist Oskar Morgenstern the first book on game theory.
Typical of von Neumann's eclectic interests, he wanted to figure out a scientific approach to bluffing in poker. In poker, you don't know what cards the other players have, giving players the opportunity to misrepresent strengths and weaknesses.
In other words, you make decisions not only on incomplete information, but also on information deliberately mis-represented. Von Neumann realized that how best to act on these situations would apply to much decision-making in business, and society in general.
Game theory eventually became useful accepted methodology in fields such as economics, political science, and biology. Today, game theory offers insights into practical pursuits, including business strategy.
As people in real life don't behave according to mathematical theory, game theory is not necessarily a predictive tool. Nevertheless, it can guide decisions, and is a useful analytical device for explaining what happens in many real world situations. In particular, it explains why results among multiple participants often end up less than optimal for an individual, and the combined result inferior to what could have been attained for the group.
In 1950, two researchers at the RAND Corporation, a prominent think tank, produced a paper showing mathematically that final solutions are often less than optimal. A mathematician at Princeton University, Albert Tucker, devised a simple story to illustrate this phenomenon, "the prisoners' dilemma."
Smith and Jones are arrested for a crime for which the evidence is murky and inconclusive. If both deny, they get off with a lesser charger, perhaps no more than a fine. If both confess, they share the rap, and do moderate prison time.
They are held separately. Smith is offered a deal - if he confesses, and implicates, i.e., rats on Jones, while Jones denies, Smith goes free. Jones takes the rap and does heavy jail time. Jones is offered the same deal - if he confesses and rats on Smith, and Smith denies, Jones goes free, and Smith does heavy jail time.
To repeat, if both confess, they do moderate jail time. If both deny, they get off easily because of the murky evidence.
So, isn't it obvious that they should both deny? A light fine would be the optimal solution for both combined. But it's not likely to happen.
Smith considers his options under alternative scenarios:
Suppose Jones denies. If Smith denies, Smith gets off lightly, paying a fine. But if Jones denies and Smith confesses, Smith goes free and Jones takes the entire rap, doing heavy jail time. Any honor among thieves, sentiment, morality, or sense of loyalty apart, if Jones denies, Smith has the incentive to confess. But he has no assurance that Jones will deny.
Smith has to consider the possibility, actually, the likelihood, that Jones will confess.
If Jones confesses while Smith denies, Smith takes the entire rap and serves heavy jail time - worst-case scenario for Smith, while Jones goes free.
If both Jones and Smith confess, they share the rap and do moderate jail time. Again, Smith has the incentive to confess. Better to do moderate jail time than to play the sucker, take the full rap, and do heavy jail time.
Note that whatever Jones might do, confess or deny, Smith has the incentive to confess.
Since Jones is offered the same deal, his reasoning is exactly the same. No matter what Smith does, confess or deny, Jones has the incentive to confess.
Net result: Both confess, and both end up serving moderate jail time even though their optimal solution would have been for both to deny. But because neither knows what the other will do, and both have the incentive to avoid worst case scenario - playing the sucker and doing heavy jail time while the other goes free - a less-than-optimal solution results for both.
This "game" gets ever so more complicated and less likely to result in optimal solution when there are more players involved. And it flies in the face of the proposition that each individual acting in his/her own self-interest automatically furthers the greater good.
But surely, the skeptical reader will reason, this "game" drummed up by some ivory tower dreamer has no basis in reality. Surely, rational, intelligent, honorable people would avoid getting into this dilemma - this trap.
Really? Even when there are billions of dollars in profits at stake? So huge Wall Street financial institutions run by super ambitious, clever people with egos big enough to fill an auditorium are about integrity, honor and putting the economic health of the nation above billions of dollars in profits? And individual traders are willing to subject multimillion-dollar bonuses to the broader objectives of ethics, integrity, and a healthy economy? If you believe that, I have a bridge in Brooklyn about which we should discuss.
Wall Street is about making money - lots of it, period. So what if the scions of Wall Street have to invent fuzzy products and new schemes to make commissions, stretch the law, bribe the congress to eliminate regulations, or pass laws unleashing them from constraints? Let someone else worry about the economy.
Next week: Wall Street, the sub-prime mortgage fiasco, and the prisoners' dilemma.
- John Waelti of Monroe can be reached at jjwaelti1@tds.net. His column appears each Friday in The Monroe Times.
