Revisiting Reif and Schmitt: Second-Order Effects in European Parliament Elections, 1979-2019

Are elections to the European Parliament (EP) truly “second-order” reactions to “first-order” national elections and politics? Ever since Reif and Schmitt (1980), the consensus in both academic and journalistic accounts has been a resounding yes. However, comprehensive quantitative confirmation of this consensus has been lacking, and those studies that have been done have been temporally limited, regionally limited, or at least somewhat critical of Reif and Schmitt’s observations. After setting the stage with a historical account of the EP’s creation and evolution, I attempt to fill this lacuna by constructing a new data set of party-election observations (over 1300 in total) from all national-level EP elections (1979-2019, including special elections in new member states). Using the data and the six dimensions of second-order elections noted by Reif and Schmitt, I dissect various variables (turnout, structure, party results, etc.) that may be different between EP and national general election, followed by the construction and testing of an overall model of EP elections as second-order elections. I then proceed to detail possible implications of the findings and avenues for future research