Pedro Magalhães

Elections and accountability: is retrospective voting blind or myopic?

This is the title of a panel I will be proposing for the ECPR General Conference in Oslo, 6-9 September 2017. It’s part of a Section entitled “Democracy for Realists” in a Comparative Perspective, chaired by Hanna Wass and Kees Aarts, which seeks to assess the generalizability of the main findings of Achen and Bartels’s Democracy for Realists: Why Elections Do Not Produce Responsive Government. The abstract of the panel is below. Looking forward to presenting a paper that fits this research agenda? I’ll be happy to consider it for the panel proposal (deadline: February 15th). Get in touch:

Abstract: Representative democracies involve delegation from citizens to political actors. In highly complex contexts, where a lot of knowledge is required to evaluate issues and correctly assign responsibility for outcomes, such delegation is potentially fraught with problems. It may become difficult to select qualified and properly motivated agents, or to avoid that those who are selected pursue their own agenda or private gain rather than the welfare of citizens. But in systems with free and fair elections, retrospective voting provides a solution to properly motivate agents, rewarding incumbents for favorable outcomes and punishing them for bad ones. All this requires is that voters behave in elections as competent evaluators of the past performance of incumbents.

But do they, and can they? An increasing amount of research has suggested that voters tend to give disproportionate importance to very recent, rather than cumulative, performance. Furthermore, their evaluations of incumbents seem to be powerfully affected by facts and events unrelated to the performance of incumbents and over which they exert no control. Finally, they seem to be vulnerable to manipulation and appeals to partisan or other group identities, deviating their attention from objective outcomes. This panel invites papers addressing these topics. How prevalent are these deviations from optimal retrospective voting? Were they more prevalent in the past than they are now? What contexts — social, economic, political, institutional — magnify these deviations? To what extent do they force us to rethink what a “well-functioning democracy” should — and can —look like?