Does electoral accountability exist under democracy? Given its normative importance, this question remains central to empirical research. Even so, scholars do not agree on a definition of the dependent variable. Some insist that in order for accountability to exist, voters must reelect incumbents for good performance or remove them from office for bad performance. However most scholars require only that voters reward or punish incumbents for good or bad performance, measured in terms of vote shares. We explore the extent to which accountability exists across the range of definitions, from vote shares through changes in seat shares to the question of incumbent survival in office. We have two main findings. First, accountability exists across the range of potential meanings of the word - including changes in partisan control. Second, institutional factors thought to mitigate connections between past performance and vote shares do not function as predicted, both across meanings of accountability and across democratic regimes. Instead, partisan change - the most substantively meaningful measure of electoral accountability - is more likely under conditions of low clarity of responsibility. More broadly, we identify regime-specific contexts to show that opportunities for elite control often diffuse or negate the effects of clarity of responsibility. These findings advance our understanding of the nature and extent of accountability linkages under different constitutional designs.
Copyright 2011 Elsevier B.V., All rights reserved.