By Steffen Zittlau and Thomas Gschwend, University of Mannheim
What is the story?
Last election in Lower Saxony was a very close race. In the evening of Jan. 20, the first projections that came in saw the incumbent right-wing CDU-FDP coalition in a very narrow lead, later in the evening the projections shifted in favor of a left-wing SPD-Greens coalition. At the end of a long election night, the CDU-FDP coalition fell short of only around 1,000 votes, giving to their opponent a one-seat majority in parliament.
A close race was predicted by the polls,
What took everybody by surprise was the performance of the FDP.
While it was lingering around the 5% electoral threshold according to most polls (see our previous blog entry), it ultimately reached 10% of the votes. How to explain these differences? This time is not so much the polls that went wrong (see another previous blog entry on polls at the last Catalan election), an explanation needs to be found in the Lower Saxony’s electoral system and the incentive it gives voters, and in some specific campaign anecdotes.
Figure 1: Results of Lower Saxony 2013 election
General explanation: Ticket-splitting
The Lower Saxony parliament, as most German states’ parliaments, is elected through a compensatory mixed-member PR system, where voters cast one vote for a district candidate, and one vote for a party list. This two-tiered system offers the opportunity to strategically split the ticket in a way that favors the preferred coalition. For example,
CDU supporters ensure a coalition with the FDP by ‘renting out’ their party list vote to the FDP, while still casting a candidate vote for the local CDU candidate.
This strategy, known in the literature as ‘rental vote’ (Leihstimme), becomes especially important if voters have reasons to expect that their preferred coalition partner will fall below the electoral threshold, as it was in the case of the FDP in Lower Saxony.
To assess the importance of this practice, Figure 1 reports the difference between the vote share obtained by the parties in the two tiers at the district-level. The respective junior coalition partners, i.e. FDP and Greens, obtained more list than candidate votes in essentially all districts, while their senior coalition partners, i.e. CDU and SPD, got more candidate votes than list votes. This strongly suggests that both camps were able to coordinate their votes. However, the CDU-FDP camp seemed to have better coordinated: On average FDP list vote shares are more than 3 times higher than their candidate vote shares. As a matter of comparison, this factor is only of 1.3 for the Greens. This is very much in line with an analysis conducted by the polling firm Infratest dimap, which estimates that the FDP received up to 104.000 list votes from CDU supporters, which represents about a third of its total list votes.
Figure 2: Difference of party list and candidate vote share (in %). Positive numbers indicate that the party has received more party list than candidate votes
Specific anecdote: The district 22
As reported in Figure 2, one district does not follow this general pattern. In the 22nd electoral district, the FDP received 17.5% of the candidate votes, and only about 7% of the party list votes. This seems to undermine our story about coalition voting, but a specific anecdote explains this result. In district 22, the CDU candidate dropped out of the race just before the election because of a scandal. The CDU therefore heavily encouraged their supporters in the district to support the FDP candidate. Nevertheless, the CDU candidate had to remain on the ballot pro forma, and still managed to gain about 22.5% of the candidate votes. In other words,
The CDU-FDP camp split their votes but failed to coordinate. As a result, the SPD candidate was able to win this district.
Ironically, if this coordination in district 22 had been successful, the FDP candidate would have won the district.
For more on ‘rental voting’ and other strategic incentives in mixed-electoral systems, see:
- Meffert, Michael F., and Thomas Gschwend. 2011. “Polls, coalition signals and strategic voting: An experimental investigation of perceptions and effects.” European Journal of Political Research 50(5): 636–667
- Meffert, Michael F., and Thomas Gschwend. 2010. “Strategic Coalition Voting: Evidence from Austria.” Electoral Studies 29(3): 339-349.
- Gschwend, Thomas. 2007. “Ticket-Splitting and Strategic Voting under Mixed Electoral Rules: Evidence from Germany.” European Journal of Political Research 46(1): 1–23.
