This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: Today in Comparative Politics • 2010 Rutgers/Ritsumeikan Spring and Summer Exchange Program • Majority rule • Single peakedness • Median voter theorem • Chaos in multidimensional problems • Arrow impossibility theorem Intransitive majority preferences • A preference relation R is transitive if XRy and yRz imply xRz. • When majority preferences cycle, • The majority preference relation is intransitive. • The outcome under a binary procedure is completely determined by the agenda. • Whoever controls the agenda has great power here. • “Anonymous” majority rule doesn’t look so egalitarian now. • Opportunities for strategic voting are numerous. Does majority rule always cycle? • Majority voting is very commonly used in distributivepolitics settings. • Saw last time how easy it is for cycles to occur in that case. • But there is an important circumstance in which majority rule is well behaved. Single peakedness • Assume that all individual preferences are strict. • Assume that there is an odd number of voters. • Suppose that in every triple of alternatives {x, y, z} there exists one alternative, say x, that every individual agrees is not worst . • That is, for all i either xP i y or xP i z • Then preferences over {x, y, z} are said to be single peaked . • Why “single peaked”? A singlepeaked preference profile y x z 1 2 3 4 If majority preferences cycle, single peakedness doesn’t hold 1 2 3 x y z y z x z x y Nonsingle peaked preference profile x y z 1 2 3 x z y 1 3 2 y x z 2 1 3 y z x 2 3 1 z x y 3 1 2 z y x 3 2 1 Black ’s theorem • If the number of voters is odd and preferences are singlepeaked, the majority preference relation is transitive. • Proof: pretty easy algebra • I will leave the argument in the Sakai slide deck marked again in yellow. Suppose x P maj y P maj z P maj x xPyPz xPzPy yPxPz yPzPx zPxPy zPyPx Suppose x P maj y P maj z P maj x xPyPz n 1 xPzPy n 2 yPxPz n 3 yPzPx n 4 zPxPy n 5 zPyPx n 6 Suppose x P maj y P maj z P maj x If x is not worst, n 4 = n 6 = 0 xPyPz n 1 xPzPy n 2 yPxPz n 3 yPzPx n 4 zPxPy n 5 zPyPx n 6 Suppose x P maj y P maj z P maj x If x is not worst, n 4 = n 6 = 0 If y is not worst, n 2 = n 5 = 0 xPyPz n 1 xPzPy n 2 yPxPz n 3 yPzPx n 4 zPxPy n 5 zPyPx n 6 Suppose x P maj y P maj z P maj x If x is not worst, n 4 = n 6 = 0 If y is not worst, n 2 = n 5 = 0 If z is not worst, n 1 = n 3 = 0 xPyPz n 1 xPzPy n 2 yPxPz n 3 yPzPx n 4 zPxPy n 5 zPyPx n 6 Suppose x P maj y P maj z P maj x If x is not worst, n 4 = n 6 = 0 If y is not worst, n 2 = n 5 = 0 If z is not worst, n 1 = n 3 = 0 xPyPz n 1 xPzPy n 2 yPxPz n 3 yPzPx n 4 zPxPy n 5 zPyPx n 6 xP maj y: n 1 + n 2 + n 5 > n 3 + n 4 + n 6 (1) Suppose x P maj y P maj z P maj x If x is not worst, n 4 = n 6 = 0 If y is not worst, n 2 = n 5 = 0 If z is not worst, n 1 = n 3 = 0 xPyPz n 1 xPzPy n 2 yPxPz n 3 yPzPx n 4 zPxPy n 5 zPyPx n 6 xP maj y: n 1 + n 2 + n 5 > n 3 + n 4 + n 6 (1) yP maj z: n 1 + n 3 + n 4 > n 2 + n 5 + n 6 (2) Suppose x P...
View
Full
Document
 Fall '09
 BLAIR
 Comparative Politics, Media

Click to edit the document details