Distribution of seats according to d’Hondt
This method to distribute seats was named after the Belgian legal academic Victor d’Hondt (1841 – 1901), who described it in a publication issued in 1882.
dʼHondt’s method was used for distributing the seats up to and including the Election to the 10th German Bundestag in 1983 and from the 1st to the 5th parliamentary term until 1969 for the distribution of seats in the committees and bodies of the German Bundestag. There it is used to date to fill the positions of some specific bodies, e. g. the Committee for the Election of Judges. The seats of the German members of the European Parliament were also allocated according to dʼHondt at the first and the second European elections in 1979 and 1984. The procedure is moreover decisive for the distribution of seats in the elections to the Länder parliaments in Niedersachsen, the Saarland and Sachsen.
It is a relatively simple method for calculating the proportional distribution of seats according to maximum numbers on the basis of the number of votes. The number of second votes obtained by each party is divided successively by 1, 2, 3 and so forth until the number of maximum numbers calculated corresponds to the number of seats to be distributed. Then the maximum numbers of the parties and thus the distribution of seats are determined. Here is an example calculation:
| Party A | 10,000 votes |
| Party B | 6,000 votes |
| Party C | 1,500 votes |
| Total | 17,500 votes |
| Party A | Party B | Party C | ||||
|---|---|---|---|---|---|---|
| Number of votes obtained | 10,000 | 6,000 | 1,500 | |||
| Divisor | maximum number | sequence of assigned seats | maximum number | sequence of assigned seats | maximum number | sequence of assigned seats |
| : 1 | 10,000 | (1) | 6,000 | (2) | 1,500 | |
| : 2 | 5,000 | (3) | 3,000 | (5) | 750 | |
| : 3 | 3,333 | (4) | 2,000 | (8) | ||
| : 4 | 2,500 | (6) | 1,500 | |||
| : 5 | 2,000 | (7) | 1,200 | |||
| : 6 | 1,667 | |||||
| Total number of seats to be allocated: | 5 | 3 | 0 | |||
The distribution of seats by maximum numbers also determines the order in which the seats are allocated, which is important for instance for who is to become chair when committee seats are distributed. If there are several identical maximum numbers when the last seat is distributed, it is assigned by lot.
With any distribution method there may be such ambiguities without clear rules specifying which of several parties with the same values (maximum number or residual) shall be entitled to a seat. The ambiguity is significant only if it occurs when the last seat is allocated, in the above example this would be the allocation of the seventh seat if there were only seven seats to be distributed. In that situation an assignment by lot has to be provided for with dʼHondt as with any other distribution method.
If the shares of the individual parties differ considerably in size, the distribution of seats according to d’Hondt may lead to larger deviations from proportionality, with smaller parties being put at a disadvantage (compare e. g. the results of the distribution of seats according to Hare/Niemeyer or Sainte-Laguë/Schepers). This is the main reason why the method according to dʼHondt was replaced by the Hare/Niemeyer method at the Election to the 11th German Bundestag in 1987.
Since 2009 Sainte-Laguë/Schepers method is used for the distribution of seats in Bundestag elections and the elections to the European Parliament.
Last update: 1 October 2014