Patrimonial Economic Voting: A Neglected Dimension . Richard Nadeau y University of Montreal Martial Foucault z University of Montreal Michael Lewis-Beck x University of Iowa December 2008 Abstract Economic voting studies have been dominated by the classic reward- punishment paradigm, where voters vote for the incumbent under good economic performance, but against under bad. This paradigm works well when the economic issue is a valence issue, such as prosperity. However, it leaves out positional economic voting, where the voters place in the eco- nomic structure inuences policy preference and, thus, party preference. More precisely, we suggest that the better the economic location of voters in terms of wealth (or patrimony), high-risk wealth in particular, the more they will vote right, because the right promises a better return on their investments. We demonstrate this e/ect in French presidential election data, from three national surveys 1988, 1995, 2002. The patrimonial e/ect well exceeds other economic e/ects tested, and does so under strong statistical controls. A rst version of this paper has been presented at the 2008 APSA Conference in Boston. We thank Paul Abramson, Sunshine Hillygus and Abel Franois for their helpful comments. y Full Professor, Department of Political Science, UniversitØ de MontrØal, richard.nadeau@umontreal.ca z Assistant Professor, Department of Political Science, UniversitØ de MontrØal, and Depart- ment of Economics, CES CNRS, UniversitØ Paris 1, martial.foucault@umontreal.ca x Full Professor, Department of Political Science, University of Iowa, michael-lewis- beck@uiowa.edu 1