Söderlund, P. (2016): “Candidate-centred electoral systems and change in incumbent vote share: A cross-national and longitudinal analysis”, European Journal of Political Research 55: 321–339

In his recent article, Söderlund (2016) studies to what extent an electoral system’s candidate-centeredness affects the diffusion of political responsibility. To answer this question, he analyses first, whether in candidate-centered electoral systems shifts in incumbent’s parties’ support are smaller, second, whether the tempering effect of candidate-centeredness on incumbent vote change has increased over time and, third, whether candidate-centered electoral systems have an conditioning effect on economic voting. The cross-sectional time-series analysis of 23 OECD countries shows that incumbent parties tend to win or lose fewer votes in candidate-centered systems: in PR systems with open list or single transferable vote, incumbents are least likely to experience shifts in vote shares as voters seem to focus on individual candidates’ characteristics more than on party performance. On the contrary, prime minister’s parties in closed list or ordered list PR systems – hence in party-centered systems – are subject to greatest shifts in their vote share. Moreover, the author demonstrates that election-to-election variability in vote shares for the prime minister’s parties have increased in party-centered systems between 1961 and 2014. However, in contrast to the first two hypotheses, evidence for the hypothesis that electoral fortunes of prime minister’s parties are stronger linked to the economic success in party-centered electoral systems is statistically only very weak. Thus, his argumentation that incumbent parties in candidate-centered systems are not punished or rewarded as much based on economic performance does not hold empirically.

This well-argued and empirically strong paper helps us to understand how electoral system design might affect the diffusion of political responsibility. It takes an interesting angle by stressing electoral systems’ intraparty dimension of competition rather than relying on the classic distinction between the three main types of electoral system - proportional representation, majoritarian and mixed systems. The inconclusive results about the conditioning effect of contextual factors on economic voting might become clearer when the micro-level is included into the analysis. At the individual level, voters might not perceive economic performance as government responsibility or at least might not believe that the government is provided with the necessary powers to be responsible for economic performance. In turn, the electorate might not punish government parties in times of economic downturn.