The problem of preferences aggregation

winner may depend on the presence of a distant loser

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: does not satisfy Independence of Irrelevant Alternatives  ­ Usually easy to manipulate (by ranking alternatives strategically...)  ­ Winner may depend on the presence of a distant loser. (Also might not choose Condorcet winner when it exists.)  ­ Strategic manipulation of the voting agenda: agenda setting  ­It is possible to collectively select Pareto ­dominated options.  ­ Any voting rule violates at least one of Arrow's axioms Example: Borda count Paths out of the impossibility. We will follow basically two routes:  ­ Remove unrestricted domain (restriction on preferences)  ­ Analyse institutional arrangements Back to Condorcet paradox: overcome if we assume that 1) the policy space is uni ­ dimensional 2) preferences are single ­peaked Social Welfare Functions (normative) W = W (u1; u2…un) Pareto improvement A change form the status quo such that nobody is harmed and at least one...
View Full Document

This document was uploaded on 03/10/2014 for the course AECO 1330A at Sciences Po.

Ask a homework question - tutors are online