GTO2-1-05: Social Choice: Arrow's Theorem

1. Arrow's Theorem and its prerequisites are introduced, establishing that there must be at least three candidates for the theorem to apply, and describing the notation for a set of strict preference orderings among agents.
2. The proof of Arrow's Theorem builds on three key conditions: Pareto efficiency, independence of irrelevant alternatives (IIA), and nondictatorship.
3. Pareto efficiency is defined as a condition where the social welfare function must agree with the unanimous preference of all agents between any two outcomes.
4. Independence of irrelevant alternatives is the principle that the social choice between two outcomes should solely rely on individual preferences for these outcomes, uninfluenced by other irrelevant alternatives.
5. Nondictatorship ensures that no single agent can dictate the social welfare function's preferences, which would otherwise render all other agents' preferences irrelevant.
6. Through the formal proof, it is demonstrated step-by-step how a social welfare function that satisfies Pareto efficiency and IIA inevitably leads to a dictatorship, contradicting the nondictatorship condition.
7. The proof establishes that the pivotal voter, referred to as n star, can single-handedly determine social rankings, which infringes upon the condition of nondictatorship.
8. The tutorial concludes by explicating why real-world voting systems commonly violate IIA to avoid undemocratic outcomes, suggesting the practical implications of Arrow's Theorem in the design of social choice mechanisms.

Key questions the transcript answers:
- What are the necessary conditions for Arrow's Theorem to apply?
- What is Pareto efficiency as it relates to social welfare functions?
- How is independence of irrelevant alternatives defined in the context of social choice?
- What does nondictatorship mean in relation to a social welfare function?
- How does the proof of Arrow's Theorem show the impossibility of a social welfare function that is both Pareto efficient and independent of irrelevant alternatives without being dictatorial?
- What are the implications of Arrow's Theorem for the design of voting systems?

Answers:
- For Arrow's Theorem to apply, there must be at least three candidates (outcomes) over which agents have strict preference orderings.
- Pareto efficiency is a condition that requires the social welfare function to align with the unanimous preference of all agents when ranking any two outcomes.
- Independence of irrelevant alternatives stipulates that the social welfare function should rank two outcomes solely based on individual preferences for these outcomes, with no impact from other unrelated options.
- Nondictatorship means that no single agent should have the power to dictate the preferences of the social welfare functions, ensuring that all agents' preferences are considered.
- The proof of Arrow's Theorem reveals that any social welfare function adhering to Pareto efficiency and independence of irrelevant alternatives ultimately confers absolute authority over social rankings to a single agent, creating a dictatorship and violating the nondictatorship condition.
- Arrow's Theorem implies that in practice, voting systems often sacrifice the independence of irrelevant alternatives to uphold democratic values and prevent a single voter from having dictatorial control over outcomes.

Core Takeaway:
The core problem described in the transcript is the challenge of creating a fair and democratic social welfare function that meets certain desirable properties: Pareto efficiency, independence of irrelevant alternatives, and nondictatorship. The consequence of not solving this problem is that any social welfare function that meets the first two conditions inherently leads to a dictatorship, negating the possibility of a democratic process that respects the preferences of all agents.

The key ideas to address this problem are:
- Accepting that while Pareto efficiency is crucial to maintain unanimity in social choices, the independence of irrelevant alternatives may need to be compromised to avoid dictatorial outcomes.
- Understanding that the inherent limitations of collective decision-making require trade-offs between different desirable properties of social welfare functions.
- Lastly, recognizing the need for designing social choice mechanisms that, while not perfect, strike a balance between respecting individual preferences and ensuring a fair collective decision-making process.