European Parliament elections are often regarded as being ‘second-order’ contests, but is this always the case? Outlining results from a new study in Denmark, Derek Beach explains that when parties run strong campaigns, EU issues are debated in a visible way, and there is a high level of media coverage, these second-order dynamics can be broken. As such, a strong showing by Eurosceptic parties in the 2019 European elections should not be dismissed as a mere protest vote – it may instead reflect voters’ underlying attitudes towards the EU.
The headlines after the 2014 European Parliament (EP) election were that it was a protest election, with many voters staying home (42% turnout), and those voters who showed up decisively turning their backs on mainstream parties. Polls suggest that the 2019 EP election might see much stronger gains for Eurosceptic parties. If this occurs, how should we interpret this? Is it due to voters using the election to punish elites, or as I suggest in the following, might it be because voters are actually taking the vote seriously and expressing their underlying attitudes towards the EU through their votes?
The most persistent and widespread claim made about how voters behave in EP elections is that voters treat them as ‘second-order’ elections in which discussions about national issues dominate. When voters feel that there is little at stake and they have little information about the issues, they will treat an election as a second-order affair that does not need to be taken seriously. In relation to the EP, Brussels is seen by many voters as being distant, while the policies they discuss are incredibly technical, and are rarely covered in the national press.
In contrast, voters treat national parliamentary elections as first-order affairs, with voting decisions based on their attitudes towards the most important issues of the day (e.g. attitudes towards economic distribution or immigration) and the performance of incumbents. When voters treat an election as second-order, they either stay home, or if they vote, they base their decision on national concerns such as attitudes towards the incumbent government and/or the desire to punish elites by voting for smaller, non-centrist parties on the left or right.
There is considerable evidence that past EP elections have been dominated by second-order dynamics, with low turnout, centrist parties doing poorly (especially those in government), and fringe parties on the left and right outperforming how they do in national parliamentary elections. However, there are exceptions to EP elections being second-order affairs. One of these cases was the 2014 EP election in Denmark. In a recent study co-authored with Kasper M. Hansen and Martin V. Larsen, we deployed a large survey instrument to research whether second-order dynamics were at play in the vote in Denmark, asking new sets of voters the same set of questions every 14 days (seven waves in total), starting from two and a half months before to election day.
We hypothesised that a substantial political campaign in which multiple candidates actually debate EU issues, and where the national media actually reports on EU issues, can result in voters no longer treating the election as second-order. A strong campaign can in theory increase the interest of voters in EU affairs and the information they possess. Taken together, this can lead to voters deciding based on their EU attitudes instead of national political allegiances or attitudes.
Evidence from our surveys suggests that campaign information increased interest in the election and knowledge about the EU, resulting in voters making decisions increasingly based on their underlying EU attitudes. First, we found that as the campaign proceeded, the Danish national media increasingly carried EU-related stories (see figure 1).
Figure 1: Percentage of newspaper stories dealing with EU across the 2014 European Parliament election campaign in Denmark
Note: Percentage calculated from media content analysis (DKOPT project), coded from a daily randomly selected page in three different large national newspapers in Denmark (Jyllands-Posten, Politiken, BT), total number of stories =2,248, EU stories = 1,237.
The increased media attention was then reflected by a rise in voter interest in the EU, as well as their awareness of their own underlying attitudes towards the EU (see figures 2 and 3). We did not find that political arguments during the campaign changed voter attitudes, but instead provided information that enabled voters to make their underlying attitudes more explicit. We tested this by re-interviewing the respondents from our first wave after the election to see whether their underlying attitudes towards EU issues had changed during the campaign.
Figure 2: Interest in EU issues across the 2014 European Parliament election campaign in Denmark
Note: The vertical lines indicate 90 pct. confidence intervals, n=2,310.
Figure 3: Percentage of voters that were able to report their EU attitudes
Note: The vertical lines indicate 90 pct. confidence intervals, n=2,310.
Taken together, increased information and interest in EU affairs meant that as the campaign progressed, voters relied more on their underlying EU issues and less on national political allegiances or national left-right attitudes when deciding which party to vote for in the 2014 EP election. Evidence for this is depicted in figure 4, where we can see that as we get closer to election day, voters increasingly relied on their EU attitudes to choose who to vote for, and less on second-order, national factors.
Figure 4: Average marginal effect of pro-integration attitudes and pro-government attitudes across the 2014 European Parliament election campaign in Denmark
Note: The vertical lines indicate 90 pct. confidence intervals. Derived from models (1) and (2) in table 3.
Our findings suggest that if EP campaigns articulate EU issues, candidates take a stand on these issues, and if the media reports on EU issues, voters then make choices based on EU issues instead of treating the election as a second-order, protest election.
These findings have implications for understanding probable outcomes at the 2019 EP elections, in which polls suggest that Eurosceptic parties on the far right and left will do even better than in 2014. Whereas EU issues had little visibility in many member states in the past, this does not hold any more. After the Eurozone and migration crises, EU issues have become highly sensitive and are now debated in many countries. In Italy, Eurosceptic parties argue that there are severe problems with the EU’s common asylum and immigration policy that need to be fixed. In Spain, unpopular austerity policies were identified with the EU. In these and many other countries, the salience of EU issues has risen significantly.
Increased media coverage and debate about EU issues in the current EP election campaign could therefore result in them ceasing to be sleepy second-order affairs. Based on our research, a strong showing by Eurosceptic parties might not be due to voters treating the EP election as a second-order protest election, but instead an expression of their underlying attitudes towards the EU. While a strengthening of Eurosceptic parties might make it more difficult for the EP to legislate on matters important to voters, it can also be seen as a positive democratic sign for the EU as a whole because voters are taking the elections seriously.
For more information, see the author’s accompanying paper in Political Science Research and Methods (co-authored with Kasper M. Hansen and Martin V. Larsen)
Note: This article gives the views of the author, not the position of EUROPP – European Politics and Policy or the London School of Economics. Featured image credit: CC BY 4.0: © European Union 2019 – Source: EP
Derek Beach – University of Aarhus
Derek Beach is a Professor in the Department of Political Science at the University of Aarhus.