Who Votes for Chavismo? Class Voting in Hugo Chavez's Venezuela
Noam Lupu, Latin American Research Review, forthcoming
Abstract: The conventional wisdom about contemporary Venezuelan politics is that class voting has become commonplace, the poor doggedly supporting Hugo Chavez while the rich oppose him. This class voting is seen as both a new feature of Venezuelan politics and a puzzle given the multiclass bases of prior populist leaders in Latin America. I clarify the concept of class voting by distinguishing between monotonic and non-monotonic associations between class and vote choice. Using survey data, I find that only in Chavez's first election in 1998 was class voting monotonic. Since then, class voting in Venezuela has been non-monotonic, with the very wealthiest Venezuelans disproportionately voting against Chavez. At the same time, Chavez's support appears to have increased most among the middle classes, not the poor. Finally, I find that whatever effect Chavez may have had on overall turnout, his efforts have not disproportionately mobilized poor voters.
State Income Inequality and Presidential Election Turnout and Outcomes
James Galbraith & Travis Hale, Social Science Quarterly, December 2008, Pages 887-901
Objective: This study examines the links among income inequality, voter turnout, and electoral choice at the state level in recent presidential elections.
Methods: We introduce two new state-level ecological data sets, estimated annual Gini coefficients of income inequality from 1969 to 2004 and a measure of income segregation across Census tracts within states in 1999. We test for associations among inequality, turnout, and party preference with cross-sectional, fixed-effects, and multilevel analyses.
Results: The cross-sectional effect of inequality on voter turnout and electoral choice is ambiguous. However, a fixed-effects analysis links higher income inequality to lower voter turnout and also to a stronger Democratic vote. Multilevel results indicate that higher levels of economic segregation likewise are associated with depressed turnout, after controlling for individual voter characteristics and for state-level income.
(Nod to Kevin L)