The discontinuous effect of economic performance on political turnover

ABSTRACT

This paper re-examines the asymmetric marginal effect of economic variables on voting. By applying the comprehensive cross-country panel data of democracies from 1975 to 2016, the empirical results demonstrate that there is a striking discontinuity in the relationship between economic performance and political turnover, in the sense that voters are sensitive to incumbents’ economic performance when the countries are during the economic prosperity but are less sensitive during the economic recession. The results are robust no matter we use the absolute growth rate or relative growth rate of the country, and no matter voters’ reaction in response to political turnover is conditional on the institutional factor or not.

KEYWORDS:

• Political turnover
• voting behaviour
• economic growth
• economic voting

Acknowledgments

We gratefully acknowledge comments and suggestions from Jan Bruckner, Editor, and the anonymous referee. Any errors or shortcomings in the article, however, are of the authors.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

¹ The detailed literature reviews on economic voting can be seen in Nannestad and Paldam (1994) and Lewis-Beck and Stegmaier (2013).

² Dix (1984) argues that the institutional factor is another important feature in determining voting behaviour. Especially in the economic voting literature, the institutional factor called the clarity of responsibility in Powell and Whitten (1993) has also gained strong support.

³ The composite Polity2 score uses the mean of the measure of civil liberties and political rights of Freedom House and the polity score of Polity IV project. This score is widely used to measure the level of democracy in the previous literature (Giuliano, Mishra, and Spilimbergo 2013; Nunn, Qian, and Wen 2021). As a robustness check, we apply the electoral democracy index from the Varieties of Democracy (V-Dem) database (Coppedge et al. 2017) and contestation and inclusiveness variables from Coppedge, Alvarez, and Maldonado (2008) to capture the strength of democracy. The empirical results are consistent with our baseline model and are shown in Table B2 of Appendix B.

⁴ As a robustness check, we sorted the political turnover into two types: turnovers with regular exit and irregular exit. We found that turnovers with regular exit are more elastic with respect to economic fluctuations, specifically during periods of economic prosperity. The detailed discussion and empirical results are reported in Table B1 of Appendix B.

⁵ Hernández and Kriesi (2016) find that the negative impact of the Great Recession is much weaker in Central and Eastern Europe.

⁶ The coding comes from the Archigos database (Goemans, Gleditsch, and Chiozza 2009).