By Fernando Feitosa Ribeiro, Intern at the Université de Montréal
What is the story?
For the past three weeks, I have been describing the particularities of the electoral system in use to elect members of the European parliament (EP) in three interesting and often forgotten countries. In particular, I focus on the possibilities given to voters to express their preference for individual candidates and intra-party competition. Last weeks, I talked about Finland and Latvia. This week, I give a snapshot of the rules of the game in Luxembourg.
The Luxembourgish electoral system for EP elections
Another variant of the party-list proportional representation (PR) will be used to elect the 6 representatives of Luxembourg for the next EP term. As in Latvia, voters have different possibilities while in the ballot booth. On the one hand, they can vote for a party list (‘en bloc‘, see Figure 1). In doing so, they actually give a vote to each candidate of the party.
Figure 1: Example of a Luxembourgish ballot (‘en bloc)
On the other hand, voters can vote directly for individual candidates. If they opt for this last option, two subtleties appear. First, they might give their support to candidates from different parties as long as they keep their number of votes lower or equal to the number of seats to be filled (‘panachage‘, see Figure 2). Second, they might vote twice for the same candidate but still as long as they keep their number of votes lower or equal to the number of seats to be filled (see Figure 3).
Figure 2: Example of a Luxembourgish ballot (‘panachage‘)
Figure 3: Example of a Luxembourgish ballot (individual candidate vote)
As in Latvia, seats are allocated on a party-basis, except that the Hagenbach-Bischoff quota is here applied. This method consists in the division of the overall number of party votes by the number of seats to be filled plus one. Once the quota is calculated, each party obtains as many seats as its overall number of party votes meets this quota. In cases of seats remaining to be filled, the d’Hondt method is used. Finally, the seats obtained by the parties are allocated to individual candidates who obtain the highest number of individual votes within their party.
Example
Imagine there are two seats to be filled in one district and two parties: The blue party that nominates two candidates (Bob, Cindy), and the yellow party that nominates two candidates as well (Carol and Sebastian).
- The individual candidates votes and the party list votes, including all of them, are first pooled. A quota is subsequently calculated in dividing this overall number by the number seats to be filled plus one.
Quota (75 pooled votes in total / (2 + 1) = 25).
- The number of each party votes is divided by the quota
Blue party: (45 party votes / 25 = 1.8)
Yellow party (30 party votes / 25 = 1.2)
- Each party is allocated a number of seats corresponding to the number obtained through this calculation (in rounding it down). In our example, the blue party obtains a seat (Quotient = 1.8), and the yellow party obtains the other seat (Quotient = 1.2).
- The votes casted for individual candidates are processed.
Blue party (45 party votes):
Bob (25 individual votes)
Cindy (20 individual votes)
Yellow party (30 party votes)
Carol (19 individual votes )
Sebastian (11 individual votes)
- Finally, the seat obtained by each party is allocated to the candidates with the highest number of individual votes. In my example, these are Bob (25 individual votes) and Carol (19 individual votes).
Conclusion
The electoral system used to elect the Luxembourgish representatives at the EP is somewhat focused on individual candidates. However, the fact that voters might cast a vote for multiple candidates across various party lists (unlike Finland) and express only positive preferences (unlike Latvia), makes this system substantially less subject to intra-party competition. Throughout these three blog entries, I tried to draw attention to the variety of electoral systems in use for European elections, and which are sometimes left aside by researchers and political observers working on the topic. Of course, these were only snapshots. Studying the functioning of these systems and their consequences for political representation would require entire chapters.
