Elections for Class A Senators will be held on September 21 (Sunday). This will concern a third of deparments, and the term will be 6 years instead of the hilariously long 9 years. Starting in 2011, Senate elections will concern half of the departments each election.

The composition on June 30, 2008 was:
23 PCF, 95 Soc, 17 Rads, 30 centrists, 159 UMP, 6 NI.
and those up for election
3 PCF, 29 Soc, 8 Rads, 4 centrists, 56 UMP, 1 NI.

New seats: 12 new seats: 1 each for the Ain, Alpes-Maritimes, Bouches-du-Rhône, Drôme, Eure-et-Loir, Haute-Garonne, Gironde, Hérault, Guyane, Polynesia. 2 new seats each for the new DOMs of Saint-Barth and Saint-Martin. In 2011, there will be 5 new seats (in Class B seats).

Electors:

Electoral college of 50,720 elected officials (49,602 in metro France). 48,453 are local councillors, 178 are MPs, 765 are regional councillors, and 1,504 are general councillors.

Elector rules:

(Commune size : Delegates)

• 1 to 499 people: 1 delegate
• 500 to 1499 people: 3 delegates
• 1500 to 2499 people: 5 delegates
• 2500 to 3499 people: 7 delegates
• 3500 to 8999 people: 15 delegates
• 9000 to 30000 people: All councillors (either 29, 33, 35, or 39 delegates depending on pop.)
• 30000 and over: All councillors + one other delegate for each 1000 inhabitants.
• Overseas seats: Elected by the Assembly of French Abroad (AFE). So, elected by a bunch of nobodies that nobody knows or bothers to vote for. Mostly a bunch of old French ladies that are bored at home.

Elections:

• Departments with 1 to 3 senators: two-round voting. Same system as legislative elections. Vendee 2004 election
• Departments with more than 4 senators: Proportional representation with highest average, no Panachage or preferential voting. See explanation below.

1. determination of the number of candidates elected in each list:
ex: 5 seats
1532 votes expressing a view
1532/5= 306.4 (quotient)
a. seat attribution using the quotient
List A: 1023 votes/306.4 (quotient)= 3.34 or 3 seats
List B: 258/306.4=0.84 or 0 seats
List C: 251/306.4=0.82 or 0 seats
b. attribution of remaining seats using highest averages
4th seat
List A: 1023 votes/(3 seats + 1 simulated seat)= 255.75
List B: 258/(0+1)=258
List C: 251/(0+1)=251
List B gets the 4th seat
5th seat
List A: 1023 votes/(3 seats + 1 simulated seat)= 255.75
List B: 258/(1+1)=129
List C: 251/(0+1)=251
List A gets the last seat
Real-life example (Jesus, I’m sounding like a math textbook): Seine-Saint-Denis 2004 election
Overseas seats use this method too.

2008 election:
Department (number of electors): Composition by group.

Ain (1563)
2 UMP
1 new seat

Aisne (1778)
2 UMP
1 vacant (RDSE)

Allier (970)
2 UMP

Alpes de Haute-Provences (459)
1 PS

Hautes-Alpes (400)
1 UMP

Alpes-Maritimes (1814)
3 UMP
1 RDSE
1 new seat

Ardèche (979)
1 PS
1 UMP

Ardennes (957)
2 UMP

Ariège (611)
1 PS

Aube (958)
1 UMP
1 NI (liberal one-man party, DVD)

Aude (1063)
2 PS

Aveyron (859)
1 UMP
1 idiot (a MPF member in the RDSE group, the most pro-European group there is)

Bouches-du-Rhône (3262)
3 PS
2 UMP
1 UC-UDF
1 PCF

Calvados (1978)
2 UMP
1 UC-UDF

Cantal (524)
2 UMP

Charente (1093)
1 UMP
1 UC-UDF

Charente-Maritime (1626)
3 UMP

Cher (871)
2 UMP

Corrèze (748)
1 UMP
1 RDSE

Corse-du-Sud (366)
1 RDSE

Haute-Corse (536)
1 RDSE

Côte-d’Or (1566)
3 UMP

Côtes-d’Armor (1628)
2 PS
1 PCF

Creuse (500)
2 PS

Dordogne (1301)
1 PS
1 UMP

Doubs (1514)
3 UMP

Drôme (1280)
2 PS
1 new seat

Eure (1721)
3 UMP

Eure-et-Loir (1229)
2 UMP
1 new seat

Finistère (2096)
3 PS
1 UMP

Gard (1688)
3 PS

Haute-Garonne (2452)
4 PS
1 new seat

Gers (755)
1 RDSE
1 UMP

Gironde (3006)
3 UMP
2 PS
1 new seat

Hérault (2094)
1 UMP
1 RDSE
1 PS
1 new seat

Ille-et-Vilaine (2317)
3 UMP
1 UC-UDF

Indre (684)
2 UMP

Belfort (356)
1 PS

Guyane (358)
1 RDSE
1 new seat

Polynesia (697)
1 UMP
1 new seat

Saint-Martin (23)
1 new seat

Saint-Barthélemy (19)
1 new seat

Wallis-et-Futuna (21)
1 UMP

French Citizens Abroad (?) This concerns only 5 of the 12 seats
4 UMP
1 PCF

The electoral system prevents a large-scale pink wave in this election and there is little chance that the PS will gain control of the Senate. In addition, some seats are decided by a small amount of votes, so division on the left or right doesn’t help.

See:

http://www.senat.fr/senateurs/elections/2008/
http://www.senat.fr/senateurs/elections//2008/brochure.pdf
http://www.senat.fr/senateurs/elections/2008/senatoriales2008/ar/liste_composition_par_groupe_politique_serie.pdf

I didn’t follow the race as I should’ve, and was caught by surprise this Monday morning with a snippet on Le Monde about the by-election. The first round was held on Sunday the 7th, and the runoff is this Sunday. I didn’t keep tabs on the candidate, so I didn’t spot the MoDem not running a candidate.

Despite what the media said, the results are actually pretty close. On paper and strictly speaking, not really, but when you analyse them, they are. Gorges (UMP) won 47.76% of the vote. Most of that vote is UMP, of course, but a big part of the traditional MoDem vote went with him. The PS candidate won 28.09%, and a dissident took 14.51. The UMP is up 11.5% from the first by-election, so it corresponds to a chunk of the MoDem vote (which represented 18.5% last time). The PS is statistically down nearly 10 points, but the PS+PSd vote is actually up 4.6%. The rest of the MoDem vote went with one of the two PS candidates. The FN won 3.79, which is down from both the 2007 and the first by-election result (-0.39). Jamais deux sans trois. The LCR ran a candidate this time ’round, and he won 3.41%, up from 2.47 in 2007 (probably took most of the 2007 LO vote, which was about 1.2% IIRC). The Commies lost 1.4% of their by-election result and fell to 2.44. The POI (the new name for the Workers’ Party, a tiny joke Trot sect) won only one vote.

All this nice stuff gives up Right 47.76% vs. Mainstream Left and Trots 48.45% vs. FN 4.18%. Which corresponds to a very tight result. Tighter than the Rhone-11 by-election also covered here. I’m predicting a narrow UMP victory for now, and I’m eyeing a replay of the June 2007 results, where Gorges won by a handful of votes (50ish IIRC). A PS win is also a very realistic possibility, though.

A UMP re-gain would certainly be a bit of good news for the party, which of course hasn’t had the best fortunes lately. That could also be some bad news for the PS, although they might not notice it since they seem busy fighting each other this moment.

For anyone interested, the prefecture has detailed cantonal and communal data here. In PDF format.