Mood:
d'ohNow Playing: CNN
Topic: Social Choice
It has been shown previously that an optimal way to vote for one position such as President is to use Range-Approval Hybrid Voting. This is a method in which each individual rates all the candidates on some scale, and the system normalizes all votes in such a way as to guarantee that each vote has maximum effectiveness. This makes it impossible for any voter to gain a strategical advantage by misrepresenting his or her sincere ratings of the candidates. I have stated that this model would apply to economic systems as well. However, an economic system is analagous to a political system in which at the simplest level each voter votes over the field of candidates and then two candidates are elected as Presidents. Then all citizens who prefer President A would be governed by President A and all citizens who prefer President B would be governed by President B. This could be extended to an arbitrary number of Presidents.
In an economic system there are many different outcomes with each citizen getting the most preferable outcome of those available. Ultimately, there would be as many outcomes as there are participants. This is analagous to a political system with many different Presidents and each voter governed by the President that he finds most preferable. This is not to say that voters or economic participants would get the outcome that is most preferred by them. But of all the possible outcomes, each individual’s satisfaction level should increase as the number of outcomes proliferates.
Considering the Range-Approval Hybrid model in which each participant’s ratings are divided into two groups by a threshold, this would only be applicable in the case where the outcome applied to all voters. In the case where there were two outcomes and each participant could have the one best for him, then it stands to reason that the threhold should be higher. The individual voter can afford to be more choosy. Quantitatively, this formula needs to be worked out. As the number of possible outcomes gets very large, the threshold will get very close to the participant’s first choice. Therefore, this model may or may not be realistic under these circumstances. Also there are constraints in the economic model that don’t necessarily apply to the political model. For instance, the outputs of the labor economy are inputs to the consumer economy so consumer preferences need to match up with workers’ preferences for the type of work that they do. For example, not everyone can work in the canned food industry if there is a limited demand for canned foods and a certain demand for frozen foods with no willing workers in that field. The demands for labor have to match up with the demands for consumption.
This non-linearity in the economic model mitigates against any coherent strategy on the part of the participants and so they will have more incentives to represent their preferences sincerely. In my Algorithm for Employee Shift Choice, employees state their preferences about shift choice and pay. Then those preferences are considered in such a way as to give each employee their highest preference insofar as possible. The strategies available have more to do with the actual algorithm used than they do with the overall model.
