"The Strategy of Voting" (Dixit & Nalebuff, Thinking Strategically, Ch 10)

Thesis

Voters often have an incentive to misrepresent their preferences and no voting system can perfectly turn individual preferences into the will of the people. So the structure of the vote, especially the agenda order, can decide the outcome

Arrow's Impossibility Theorem

• Unrestricted Domain (Universal Domain)

• Pareto Efficiency (Unanimity)

• Independence of Irrelevant Alternatives (IIA)

• Non-Dictatorship

2 effects a vote can have

1. instrumental (changed outcome)

2. voice (changed margin)

3 cases showing the median voter has no reason to lie

1. a voter left of the median

2. a voter at the median

3. a voter right of the median

truthful voting is the dominant strategy

3 court procedures and their outcomes

1. status quo (decide guilt first) —> acquittal

2. roman tradition (start with the most serious punishment and get less harsh) —> death

3. mandatory sentencing (fix the sentence first) —> life imprisonment

When does a vote matter?

When it creates or breaks a tie

Zeno’s paradox of the heap

One vote is like one grain of sand

Mean voting (averaging)

Gives people a reason to exaggerate

Median voting

uses direction, not intensity

Condorcet paradox

Majority voting can produce self-contradictory preferences

Line-item veto

The president may be better off without one

Link to textbook

Challenges rousseau’s general will and wolff ch. 3 on voting

Free rider logic of voting — people don’t feel like its worth it to vote