DISTRICT MAGNITUDE, ECONOMIC-PERFORMANCE, AND PARTY-SYSTEM FRAGMENTATION IN 5 LATIN-AMERICAN COUNTRIES

This article makes a fresh start in the attempt to explain the number of parties in patty systems. It develops a simultaneous equations mode l to differentiate between the psychological and mechanical effects of district magnitude on party-system fragmentation. Both effects are st atistically significant and approximately equal. However, neither effe ct is very large in comparison to underlying patterns of politicizatio n, which are argued to be reflections of the number of political cleav ages in society. These cleavages predispose each party system to conve rge toward a country-specific effective number of parties within 5 ele ctions, regardless of the initial level of fragmentation, barring outs ide disturbances. Major devaluations may act as such disturbances, but the evidence so far is inconclusive. The analysis is based on new dat a from 62 elections in Argentina, Bolivia, Colombia, Peru, and Venezue la, supplemented by 30+ additional elections in Brazil, Chile, Costs R ica, Ecuador, Mexico, and Uruguay for the exploration of economic impa cts.