The body that we now know as the European Parliament (EP) first met in March 1958; at the time, it was called the European Parliamentary Assembly. It began calling itself the EP in 1962, well before the first popular election in 1979. It has officially been the EP only as of 1987 with the Single European Act. Shortly after the first European parliamentary election (EPE), political scientists Karlheinz Reif and Herrmann Schmitt analysed its electoral behaviour and clearly evaluated the EPE as a second-order election.
Almost 20 years on, in 1997, Reif took stock again. By his criteria, the 4th EPE, which took place in 1994, was still clearly a second-order election. It had low turnout; the outcome was strongly related to the popularity of national parties in particular countries; the election did not revolve around specific European issues nor did it have visible European candidates; and it wasn’t related to salient European events. There was no real electoral campaign and the cyclical pattern of voters protesting against parties governing nationally, was still clearly in place.
We are now 40 years on. I understand the EPE as an election in transition: not second order any more, but not quite first order yet either. This transition is slow. Its precondition is significant habitual and cultural change, and such change does not happen overnight, nevertheless, it happens. What is the evidence?
Europe as a political issue, as well as European policies, has vastly grown in importance over the last couple of years. There was the sovereign debt crisis and the European role in solving it. There is an ongoing discussion among the member states and in the EU institutions on immigration policy, fair sharing of related burdens, and common border protection. There is talk about a fiscal union because of the common currency. This extends to a discussion about a common labour market and welfare policy. Though opposed widely, these policy issues are genuinely European.
Furthermore, no event made Europe such a topic of public, political, and media discussion as Brexit. Next in line are the trade disputes with a US led by President Trump. The latter clearly demonstrates how important the EU is for all of its members: not only the smaller, but also the larger ones, for example on international trade issues. Disputes with the Polish and Hungarian government on policies contradicting core European values make headlines as well.
On the other hand, there are figures like French President Macron with a very strong pro-European agenda and many ideas on how to change not only individual European policies, but also the whole way the EU is taking on responsibility for the social and political wellbeing of its residents. These developments are a sign of an EU struggling with its role and identity. As for the EPE, this vote is not the second-order election it used to be 20 and 40 years ago.
Of course, none of these developments, most of which took place after the 2014 European election, have so far been analysed with respect to electoral behaviour. So what is the evidence here? It is mixed, though in line with my claim that the EPE is an election in transition. Different authors, however, arrive at different conclusions.
Analysing the 2009 EPE in Germany, Heiko Giebler and Aiko Wagner argue that its second-orderness is dwindling. Governing parties in Germany did not get punished. The state of the national economy was important in the national election of that year, but played no role for voters in the European election. European candidates where clearly more important than national ones (though the national election still was perceived as clearly more important than the EPE). Voting “with the heart,” something clearly ascribed to second-order elections by the respective theories, was not more prominent in the 2009 EPE then in the 2009 national election.
Hobolt, while analysing the 2014 EPE, argues that the issues relevant for this election were European, not national issues. She found some important issues distinctly European in nature: immigration, EU redistribution, EU unification, and EU fiscal integration. This evidence does not justify a classification of the 2014 EPE as clearly second-order. Ironically, the surge in popularity for eurosceptic political parties helped make European elections more salient.
Of course, there are more cautious perspectives as well. Looking at the role of the Spitzenkandidaten (front-runner candidates), the relationship between media and voters, as well as electoral behaviour in a number of countries, most articles in an issue of Politics and Governance in early 2016 concluded in favour of a second-order classification.
In my opinion, the upcoming EPE will not be a first-order election. When asking, however, whether it clearly is a second-order election, the correct answer is very likely no as well. My conclusion: the EPE of 2019 will be a next step in the transition of European elections from second to first order. Analyses of electoral behaviour after the election will show how far it has come.
 Reif, K. & Schmitt, H. (1980). Nine Second-Order National Elections. A Conceptual Framework for the Analysis of European Election Results. European Journal of Political Research 8: 3–44.
 Reif, K. (1997). European elections as member state second-order elections revisited. European Journal of Political Research 31: 115–124.
 Giebler, H. & Wagner, A. (2015). Contrasting first- and second-order electoral behaviour: determinants of individual party choice in the European and German Federal elections. German Politics 24(1): 46–66. doi: 10.1080/09644008.2014.949684.
 Hobolt, S. B. (2015). The 2014 European Parliament Elections: Divided in Unity? JCMS: Journal of Common Market Studies 53(3): 6–21. doi: 10.1111/jcms.12264.
 For an overview of the issue see van der Brug, W., Gattermann, K. & Vreese, C. H. de (2016). Introduction: How Different Were the European Elections of 2014? Politics and Governance 4(1): 1. doi: 10.17645/pag.v4i1.591.