Split Your Ticket? Split Your Friends: Social Network Disagreement and Partisan Preferences in the 1990 German Electionon 5 March 2018
By Debra Leiter
In September 2017, Germans elected 12 members of the Bundestag that have no living memory of a divided Germany, having been born during or even after Germany’s reunification. As of February 5 of this year, the Berlin Wall has been gone for more days that it stood. And Germans cast their first ballots as a unified Germany a little more than 27 years ago.
Yet as reunification approaches the three decade mark, there are still many lessons we can learn from this historic and unique election. By returning to the 1990 election, we can discover important distinctions that exist between voters in brand new versus established democracies. In a recent paper, I examine how much partisan preferences versus social networks influence split ticket voting amongst East and West Germans. Because we have “experienced” West Germans voting in the identical election to “new” East German voters, this election can tell us a lot about the difference in decision-making across these types of voters.
Why split ticket voting? Although it has grown in frequency, split ticket voting remains the exception, rather than the norm, for most voters. In 1990, more than 85% of West Germans and 75% of East Germans cast a unified party ballot (i.e. voted for just one party), even though the German mixed electoral system allows voters to cast two different votes for the Bundestag. Why is split ticket voting relatively less common? In order to cast a split ticket, voters have to be both motivated to support multiple parties, and have additional information necessary to do so. The strength of most individual’s partisanship and the cost of acquiring more information are enough of a barrier for most voters to support just a single party. These additional requirements of split ticket voting thus tell us something important about what motivates voting decisions, and under what conditions certain sources of influence will matter more.
We know that social networks – the interpersonal connections of friends, family, coworkers, and acquaintances – play an important role in political decision making. We learn from and are influenced by those with whom we speak and interact. A casual chat with our neighbor about the upcoming election, a fiery argument with a brother-in-law about politics, or a phone call to your friend where she mentions a campaign rally are all part of our personal election context, and shape what we know and what we think about our political decisions. In particular, disagreement within your network – not sharing party preferences with those you speak with – can play a critical role in shaping a decision to split your ticket. Not only do voters with friends from other parties learn about the relative merit of other parties besides their most preferred, but disagreement in a network can decrease your electoral certainty, causing you to split a ticket for some electoral insurance- at least one party you supported will be likely to be successful.
Yet for those who have lived life-long in a democracy, social networks compete with our own partisan preferences and political experience. Partisanship plays a strong and influential role in split ticket decisions- those with strong ties to a party have very little chance of supporting multiple parties. And political experience only increases certainty and partisan strength. Thus voters in established democracies make decisions in a competitive information environment, with social networks as one source of influence amongst many, and this may mute their overall impact.
For voters in brand new democracies, however, these partisan preferences have had little time to develop, nor do they have experience with the political system. These brand new voters cannot turn back to the previous elections to see how they voted, or who was likely to win. But voters in new democracies still have their friends and family to talk politics with, and the massive political changes are likely to make politics a hot topic of discussion. Thus, the absence of political experience and strong party ties may cause network characteristics to profoundly influence decision-making in first elections.
Using a unique survey given right before the 1990 German national election that asked respondents not just their political attitudes and preferences, but also the preferences of their five friends, I examine what motivated voters to split their ticket.
Two important stories emerge from the results. The first is that party affiliation is a stronger motivator for West German ticket splitters than East Germans. West German supporters of the two largest parties in Germany – the Christian Democrats and the Social Democrats – have only a 3% probability of splitting their ticket; for East Germans, supporters of major parties are four times as likely to split their tickets. The strength of party ties and the incentives associated with those ties are generally weaker for East Germans in this election.
Secondly, we see the clear impact of network disagreement on split ticket voting, especially for East Germans. An East German amongst only those who share their partisanship have less than 1 in 5 chance of splitting their ticket. However, if all five of their friends support a different party, the probability that an East German will split her ticket jumps by almost 50%. The same change for a West German – sharing partisanship with all your friends to of them – only increases her chance of splitting a ticket by about 16%. Even if all of their friends disagree with them, West Germans, on average, are still unlikely to split their ticket, but for East Germans, the network drastically shifts their political decision making.
Thus electoral experience matters for how our social context motivates our decision-making. By re-examining an almost thirty year old election, we reveal something important about how voters make decisions. And it’s worth noting that it is not just in new democracies where we should expect social networks to play a profound role. In a time of declining partisanship, major changes in the party systems across Europe, and increasingly competitive and unpredictable elections, the influence of social networks on political decision may certainly increase. As voters turn increasingly to their networks to help them make decisions, the friends we choose and the discussions we have will matter more and more, and in turn, may profoundly shape election outcomes.
Debra Leiter is an Assistant Professor of Political Science at the University of Missouri-Kansas City. Her research focuses on the intersection of elections, voters, and political parties, with an emphasis on Western European electorates. Her article ‘Social Networks, Predispositions and Split Ticket Voting: The Case of the 1990 German Reunification Election’ is on early view in Political Studies now. She tweets @DebraLeiter
Image: Mike CC BY-NC-ND