Majority Voting

Today a student asked if trading votes can solve the problem of inefficient majority voting. I said that trading votes may sometimes result in better outcomes, and sometimes result in worse outcomes. Here are some examples to illustrate that.

Suppose a project costs $600 and three people will each have to pay $200 for the project. Person 1 values it at 300,person 2 values it at 75 and person 3 values it at 175. From society's perspective, the project should be voted down, and with no vote trading, it would be. Person 2 would vote no and so would person 3. But person 1 can pay person 3 up to $100 and change his vote. This would turn an efficient vote into an inefficient one.

I think, that you can come up with many scenarios that sometimes improve efficiency when votes are traded and sometimes decrease efficiency when votes are traded. Because you can come up with scenarios on both sides, it is not logically necessary for vote trading to improve efficiency.

However, I do have to say that if vote trading came with zero transactions costs, then the Coase Theorem would seem to apply and the efficient solution would result. I think the question hinges on the restrictions and conditions you put on how votes can be traded. In the example above, that was only one way the votes could be traded. Notice that person 2 would be willing to pay up to $125 to ensure a no vote. this would result in person 2 avoiding the tax of $200. We never specified how votes could be traded, and without this specification, it's hard to draw any concrete conclusions.

please feel free to share your thoughts on this with Mr. Goldenballs :)