Belgian elections 25th May, 2014

[palandio]

Belgian elections are also interesting from the arithmetic view because of d'Hondt voting in districts. I will try to give my take on each race in the hope that people who know more than me will correct me:

Bruxelles/Brussel (15 seats):
Because of the large number of contending parties the effective quota will be at ca. 5%, far below the theoretical 6.25% (=1/(15+1)). That means for example that a party with between 5% and 10% will most likely get 1 seat.
In 2010 the major Dutch-speaking parties together got only 11%, the strongest being Open VLD at 2.3%. Because of this I think Groen did the only reasonable thing when allying with Ecolo in exchange for the third seat on the Ecolo list. The other Dutch-speaking parties should really think about electoral cartels, too.
If they were able to hold onto their 2010 results, MR and PS would have 4 seats each safe, with a 5th seat very likely, CDH and Ecolo 2 seats each. The rest would go to PVDA-PTB, FDF and Parti Populaire. Because of negative trends I wouldn't consider the 5th PS and MR and the 2nd CDH seat safe, though. (The 2nd Ecolo seat should be safe because of the cartel with Groen.) Other Dutch-speaking parties being involved in cartels would change the arithmetics.

[palandio]

Flemish voting districts

Antwerpen (24 seats):
The theoretical quota is 4.0%, more likely is something in the range of 3.6-3.8%.
Under this assumption the 4th CD&V seat, the 3th Open VLD seat and the 2nd Groen seat are very unsafe in the case of even slight declines. (Groen will likely grow slightly, though.) The 3rd sp.a seat is safe even against a loss of 3 percentage points. If VB loses a considerable share, as expected, they will lose their 4th seat. Antwerpen is also one of the best pick-up opportunities for the PVDA. The N-VA seat share is the most luck-dependent under the d'Hondt system.

Limburg (12 seats):
The theoretical quota is 7.7%, the effective quota will probably be in the 6-7% range.
The 3rd CD&V seat is very unsafe, as is the 2nd VB seat. On the other hand the 2nd sp.a seat and the one Open VLD seat are ultra-safe. Being there only 12 seats, PVDA and Groen will get at most one seat each.

Eastern Flanders (20 seats):
The theoretical quota is 4.76%, the effective quota will probably be something like 4.1-4.5%.
Under this assumption the 4th Open VLD seat is very unsafe, as is the 3rd VB seat. The 3rd CD&V seat is safe. The 3rd sp.a seat is relatively safe against minor percentage losses, but something like -1.5% could become dangerous. The Groen seat is very safe, with chances for a 2nd seat in the case of gains.

Flemish Brabant (15 seats):
The theoretical quota is 6.25%, the effective quota we be around 5.5%. What makes predictions difficult is the dissolution of the B.H.V district. The most likely effect is that French-speaking voters of MR, PS, CDH and Ecolo will switch to Open VLD, sp.a, CD&V and Groen respectively. Under this assumption the 3rd CD&V and particularly the 3rd Open VLD seat become much safer than before, though still at risk. The 2nd sp.a. seat is ultra-safe and ex-PS voters even provide a small pick-up chance for a 3rd seat. The single Groen seat is ultra-safe as well and with ex-Ecolo voters there is even a small chance for a 2nd seat. The 2nd VB seat is almost surely gone.

Western Flanders (16 seats):
The theoretical quota is 5,9%, the effective quota probably something in the 4.7-5.5% range. Under this assumption the N-VA has good chances to gain a 4th seat even with minor percentage gains. The 4th CD&V seat is relatively safe. The 2nd Open VLD seat and the single VB seat are very safe. The 3rd sp.a. seat is strongly at risk. The Groen has almost no chance to gain a 2nd seat. Then there's LDD, not heared much about them lately, might lose their seat, probably to the N-VA.

Altogether:
The arithmetic effect on the N-VA is the most difficult to predict.
The CD&V is at risk in Antwerpen and Limburg and needs a bit of luck and ex-CDH votes in Flemish Brabant.
The sp.a is at risk in Western Flanders.
Open VLD is at risk in Antwerpen and Eastern Flanders and needs a bit of luck and ex-MR votes in Flemish Brabant.
VB is at risk in Antwerpen, Limburg, Eastern Flanders and Flemish Brabant.
Groen has pick-up opportunities in Limburg, Eastern Flanders and Flemish Brabant, of course only if they increase their vote share.
PVDA chances depend on the effective quota which depends mainly on the number of seats. Antwerpen is of course the easiest gain, Limburg the most difficult.