Introduction

1The number of municipal councillors in the 19 municipalities of Brussels is increasing. Today – after the 2012 elections – 685 councillors are in office. In 2006, 663 were appointed and in 2000 653. After the municipal elections of 2018 the number of councillors will amount to 693. The total number of aldermen is also following this upward trend, increasing from 132 aldermen in 2000 to 141 aldermen in 2018. Those numbers just evolve according to the population size. In Brussels, the exact same rules apply as in Flanders and Wallonia. But because the municipalities of the Brussels-Capital Region are all medium-sized to very large, and because the population size of all municipalities is increasing, the number of municipal councillors and aldermen is high and continues to increase.

2In the Brussels-Capital Region, both the status of the municipalities (the powers they can exercise) and the number of municipalities have been the object of debate for a very long time now [Rowies, 1980; Witte et al., 2003; Vaesen, 2008]. Most of the time, Dutch speaking and French speaking parties take different positions in the matter. Dutch speaking parties often call for a limitation on the number of municipalities – even reducing them to a single one – and so for a strengthening of the regional institutions. These are also the institutions where a Dutch speaking presence in the council is the largest. French speaking parties defend the existing situation, with arguments that refer to the proximity of the municipal level for the citizens, and also to the institutional compromise which, in exchange for a strong protection of the Dutch speaking minority in the Region, maintained the municipalities as a political level. The fear for a blocking of the regional institutions by the right of veto for the Dutch speaking representatives is another reason to maintain the municipalities as a fully-fledged level of decision-making [Nassaux, 2018].

3In 2017, Rudi Vervoort, Minister-President of the Brussels-Capital Region, suggested to reduce the number of municipal councillors and aldermen in Brussels within the existing institutional framework (without modifying the number of municipalities or to their powers). This would address the criticism that in Brussels for a relatively limited population and for a relatively small area there would be more representatives than in the two other regions and than in other cities. A comparison of the number of representatives in Brussels with other cities at home and abroad, and also with cities which are both capital and region in a Federation (such as Vienna and Berlin), indeed shows that the number of Brussels’ representatives - especially the executive representatives - is relatively high [Bürger, 2013; Martin, 2013; van Haute, Deschouwer, et al. 2018].

4However, reducing the number of representatives is much more than just a mathematical process. The number of available seats at the elections has a direct impact on, among other things, the number of seats that can be conquered for each list, and thus for the composition of the lists, for the opportunity for specific groups to be elected and for the way in which representatives can exercise their function [e.g. Pilet, Freire and Costa, 2012]. In this contribution we want to examine the potential impact of a reduction in the number of municipal councillors and aldermen in the Brussels-Capital Region. We examine three scenarios: a reduction in the number of councillors and aldermen with 10, 20 and 30 %. These are obviously arbitrary percentages, but they allow us to figure out how the effects of a reduction in the number of representatives get more tangible as that reduction continues. For each of these scenarios, we look at the potential impact on the ideological diversity, on the workload, on the formation of coalitions, on the gender balances and on the position of the Dutch speaking elected representatives. It shows that a reduction in the number of councillors and aldermen would not be a politically neutral process. The smaller political groups would be affected, and it would also decrease the number of Dutch speaking representatives. And it is expected that decreasing the number of councillors in the BCR can also affect the gender balances.

5Thus, we also show that the Brussels context makes it very difficult to look at the institutions, the rules and the procedures separately, and to believe that they could be or become the product of a neutral and objective endeavour. The Brussels institutions are the result of a subtle and complex compromise, and a delicate interconnected framework. The most recent illustration of it is the attempt to limit the cumulation of mandates on the municipal and regional level. It led to differences between French speakers and Dutch speakers and to a conflict of interest that was invoked by the Flemish Parliament. So a “simple” reduction of the number of councillors or aldermen is not likely to be an easy task.

1. The number of representatives

Table 1. The current rules regarding the number of councillors and aldermen

Population	Number of councillors	Number of aldermen
< 1 000	7	2
1 000 – 1 999	9	3
2 000 – 2 999	11	3
3 000 – 3 999	13	3
4 000 – 4 999	15	3
5 000 – 6 999	17	4
7 000 – 8 999	19	4
9 000 – 9 999	21	4
10 000 – 11 999	21	5
12 000 – 14 999	23	5
15 000 – 19 999	25	5
20 000 – 24 999	27	6
25 000 – 39 999	29	6
30 000 – 34 999	31	7
35 000 – 39 999	33	7
40 000 – 49 999	35	7
50 000 – 59 999	37	8
60 000 – 69 999	39	8
70 000 – 79 999	41	8
80 000 – 89 999	43	8
90 000 – 99 999	45	8
100 000 – 149 999	47	9
150 000 – 199 999	49	9
200 000 – 249 999	51	10
250 000 – 299 999	53	10
> 300 000	55	10

Source: articles 8 and 16 of the NLC

8All in all, the number of councillors will increase with 52 between 2000 and 2025 (or an increase of 8 %), spread as follows:

• No increase in two municipalities: Schaerbeek and Watermael-Boitsfort

• Two additional councillors in 9 municipalities: Auderghem, Forest, Ixelles, Koekelberg, Saint-Gilles, Saint-Josse-Ten-Noode, Uccle, Woluwe-Saint-Lambert and Woluwe-Saint-Pierre

• Four additional councillors in 7 municipalities: Anderlecht, Berchem-Sainte-Agatha, Brussels-City, Etterbeek, Evere, Ganshoren and Jette.

• Six additional councillors in Molenbeek-Saint-Jean.

9The number of aldermen is established by article 16 of the NLC. However, under certain conditions, as stated in article 279 of the NLC, the number of aldermen can be increased by one in the Brussels-Capital Region. That is the mechanism of the additional alderman. On the municipal level in Brussels there is no guarantee for the participation of Dutch speaking elected representatives in the Council, such as it exists in the Region. However, since 2001 the Lombard Agreements provide for a financial reward if municipalities include a Dutch-speaking representative in the Council of aldermen or if they appoint a Dutch-speaking representative as President of the CPAS. We did not add these additional aldermen in our calculations. After all, they are the result of agreements during the formation of the coalitions in the municipalities, and therefore their number fluctuates according to the decisions per municipality as to whether or not to make use of the mechanism of the additional alderman. That does not mean, however, that in the debate on the number of municipal councillors these additional aldermen are unimportant. Quite the contrary: as the demand for a reduction of the complexity of the Brussels institutions often comes from Dutch speakers, the answer of the French speakers is that precisely that part of that complexity contributes to the protection of the Dutch speaking minority, both in the municipalities as in the regional institutions. So we simulate a linear reduction of the number of aldermen without the additional aldermen, but in the discussion on the effects we do point out the consequences on the possibility to continue applying the mechanism of additional aldermen.

10Table 3 presents the evolution of the number of aldermen during the last three legislations and a projection for the next two legislations. Between 2000 and 2012, the number of aldermen increased with one in seven councils. So the total number of aldermen in the Region increased from 132 in 2000 to 139 in 2012 (+ 5 %).

Table 3. Evolution of the number of aldermen

Municipality	2000-2006	2006-2012	2012-2018	Projection 2018	Projection 2025
Anderlecht	8	8	9	9	9
Auderghem	6	6	7	7	7
Berchem-Sainte-Agathe	5	6	6	6	6
Brussels	9	9	9	9	10
Etterbeek	7	7	7	7	8
Evere	7	7	7	7	7
Forest	7	7	8	8	8
Ganshoren	5	6	6	6	6
Ixelles	8	8	8	8	8
Jette	7	7	7	8	8
Koekelberg	5	5	6	6	6
Molenbeek-Saint-Jean	8	8	8	8	9
Saint-Gilles	7	7	7	8	8
Saint-Josse-Ten-Noode	6	6	6	6	6
Schaerbeek	9	9	9	9	9
Uccle	8	8	8	8	8
Watermael-Boitsfort	6	6	6	6	6
Woluwe-Saint-Lambert	7	7	8	8	8
Woluwe-Saint-Pierre	7	7	7	7	7
TOTAL	132	134	139	141	144

Source: [IBSA, 2016], own calculations

12The Brussels-Capital Region is fully responsible for the regulation of the local administrations. Therefore the Parliament and the Government can change the articles 9 and 16 of the NLC by means of an ordinance. But that ordinance will have to be approved with a double majority – a majority in each language group.

13We have tested three possible scenarios: a reduction in the number of councillors with 30 %, 20 % and 10 %. Every time a linear reduction was applied in each of the 19 municipalities. Table 4 presents the number of councillors for each municipality according to each of these three scenarios.

Table 4. Size of the municipal councils according to the three reduction scenarios

Municipality	Number of councillors 2012-18	- 30 %	- 20 %	- 10 %
Anderlecht	47	33	38	42
Auderghem	31	22	25	28
Berchem-Sainte-Agathe	27	19	22	24
Brussels	49	34	39	44
Etterbeek	35	24	28	31
Evere	33	23	26	30
Forest	37	26	30	33
Ganshoren	27	19	22	24
Ixelles	43	30	34	38
Jette	35	24	28	31
Koekelberg	27	19	22	24
Molenbeek-Saint-Jean	45	31	36	40
Saint-Gilles	35	24	28	31
Saint-Josse-Ten-Noode	29	20	23	26
Schaerbeek	47	33	38	42
Uccle	41	29	33	37
Watermael-Boitsfort	27	19	22	24
Woluwe-Saint-Lambert	37	26	30	33
Woluwe-Saint-Pierre	33	23	26	30
TOTAL	685	478	550	612

14The most drastic scenario (reduction of 30 %) would reduce the number of municipal councillors in the region with more than 200 (from 685 to 478). A linear 20 % reduction leads to a reduction of 135 municipal councillors and a reduction by 10 % would reduce the number of councillors with 73.

15These scenarios are of course not the only imaginable. Both smaller and larger reductions can be considered. Article 8 of the NLC stipulates a minimum of 7 councillors for the smallest municipalities (less than 1 000 inhabitants). At a reduction of 30 %, the smallest municipal councils in the BCR would have only 19 members, i.e. the number of councillors stipulated by the NLC for municipalities with between 7 000 and 8 999 inhabitants. The smallest municipality in Brussels (Koekelberg – see table 2), however, has more than 16 000 inhabitants.

16The question can also be asked whether one should stick to the principle that municipal councils should have an odd number of members. Today, this is still the case. It allows to avoid political deadlock when a municipal council is divided into two equal camps. In the case of a linear reduction of the number of councillors, the result must then be corrected if it is an even number.

17Another question concerns the evolution of the size of the municipal councils in function of the population size, as it is now stipulated in the NLC (see above). Although that mechanism stipulates a maximum number of municipal councillors for municipalities with more than 300 000 inhabitants, it evidently does not stipulate a maximum number for the whole of the BCR. In other words, if the population of the BCR and its municipalities continues to increase, the total number of municipal councillors in the region will also continue to increase.

18A possible solution is to apply the same logic as for the determination of the number of seats of the Federal Chamber or of the Walloon and Flemish Parliament. The total number of elected representatives is fixed – 150 for the Chamber – and that number is then distributed between the different constituencies according to their population size. If this rule is applied to the municipal councillors, the BCR could set a total number of councillors. That would be a absolute maximum. Evolutions in the population size by municipality would only lead to an adjustment of the distribution of seats between the municipalities. In such a scenario it would not be appropriate to simply distribute the total number of councillors over the municipalities. This would strongly disadvantage the smaller municipalities. Better is to develop a method that sets a number of councillors for the whole BCR, a fixed number by municipality (for example, 50 % of the total) and then a variable part that would be distributed over the municipalities according to the number of inhabitants. For the variable part each municipality would receive a number of councillors that corresponds to the weight of the population of the municipality in the total population of the BCR, just like it is the case in the Chamber.

19One could also possibly determine a minimum threshold for each municipality.

2. Decrease of the number of municipal councillors: possible consequences

20In the following sections, we look at the political consequences of the three scenarios (30 %, 20 %, 10 %), on the basis of the results of the 2012 municipal elections. We look at how the votes per list of those elections would be turned into seat distribution, and then also how the mandates would be distributed within the lists. Other consequences are also possible, but they cannot be illustrated just on the basis of simulations. A change in the number of available seats, for example, can affect the way in which parties draft their lists, or decide whether or not to present a list via cartel formation with other parties [Verthé and Deschouwer, 2011]. We look only to the effects that can be simulated and calculated, and know that this is not the full picture.

2.1. The political-ideological diversity

21At first we look at the extent to which a reduction in the number of municipal councillors has implications for the diversity of the political representation. Political research has elaborately demonstrated that the proportionality of the seat distribution and the number of lists that secure at least one single seat is directly related to the number of mandates to be distributed [Cox, 1997]. The fewer seats there are to distribute, the harder it is for a small list to gain at least one single seat, and the less really proportional the distribution of seats can be.

22We therefore look at the number of lists that could obtain at least one single seat in the different municipalities in 2012, and how that number would evolve if the number of councillors would be reduced with 10, 20 or 30 % (table 5). This first simulation already shows that none of the three scenarios would have a really big impact on the number of lists with elected representatives. Even a 30 % reduction would only have a limited effect. Today, the average number of lists is 5,4. That would drop to 5,0 with a decrease in the number of councillors by 30 %, and to 5,2 with a reduction by 10 or 20 %. If we check municipality by municipality, we see that in 13 of the 19 municipalities, none of the three scenarios would lead to a decrease in the number of lists with at least one single elected representative.

23There would be a loss of the only seat of Vlaams Belang in Anderlecht, of the list SAMEN in Auderghem, of the FDF list in Ganshoren (- 30 %), of the N-VA list in Jette, of the lists of FDF and ISLAM in Molenbeek (only at - 30 %) and of the PTB-PvdA list in Schaerbeek. The number of lists is evidently limited, but it has to be noted that those lists are lists which do not participate in the elections in collaboration with other lists (or cartels), and that the unilingual Dutch speaking lists would be particularly affected (see also below).

Table 5. The number of lists with at least one single seat in the municipal councils in 2012 and projections

Municipality	Number of lists with at least one single seat in 2012	- 30 %	- 20 %	- 10 %
Anderlecht	6	5	5	5
Auderghem	5	4	4	4
Berchem-Sainte-Agathe	7	7	7	7
Brussels	6	6	6	6
Etterbeek	5	5	5	5
Evere	5	5	5	5
Forest	5	5	5	5
Ganshoren	5	4	5	5
Ixelles	5	5	5	5
Jette	7	6	6	6
Koekelberg	3	3	3	3
Molenbeek-Saint-Jean	8	6	8	8
Saint-Gilles	4	4	4	4
Saint-Josse-Ten-Noode	4	4	4	4
Schaerbeek	6	5	5	5
Uccle	5	5	5	5
Watermael-Boitsfort	5	5	5	5
Woluwe-Saint-Lambert	5	5	5	5
Woluwe-Saint-Pierre	6	6	6	6
TOTAL	5,4	5,0	5,2	5,2

2.2. The workload

25The number of lists that are represented in the municipal council is not the only aspect to be looked at. We must also check if the political workload is acceptable in smaller communal councils (and small colleges of aldermen). For the municipal councils, we consider two elements: the average number of elected representatives per list and the margin of the majority controlled by the parties which, in 2012, decided to govern a municipality together.

26The first criterion – the number of elected representatives per list – is important for the work that has to be done in the municipal councils. Many and different issues are discussed during a session of the municipal council. That implies a certain distribution of responsibilities between the municipal councillors, in particular within the political groups in the council. For lists with a small number of elected representatives it is difficult, not to say impossible, to follow all those dossiers in detail. The quality control of the local government by the council would thus be directly affected.

27When the total number of municipal councillors is reduced, the number of elected representatives per list also decreases (table 6). In that case, the impact of the reduction in the number of councillors is very clear. Especially a decrease of 30 % would lead to a significant reduction of the size of the political groups. The average number of elected members per group would drop from 5,8 to 4,0. In some municipalities the drop is even more substantial. In Anderlecht, Auderghem, Schaerbeek or Uccle the average number of elected members per list would not exceed 3. That is due to the fact that often many votes go to one big list and/or that the votes are spread over a large number of small lists.

Table 6. The average number of elected representatives per list in 2012 and projections

Municipality	Average number of elected representatives per list in 2012	- 30 %	- 20 %	- 10 %
Anderlecht	4,3	3,0	3,4	3,8
Auderghem	2,8	2,0	2,3	2,5
Berchem-Sainte-Agathe	4,9	3,4	4,0	4,4
Brussels	8,9	6,2	7,1	8,0
Etterbeek	6,4	4,4	5,1	5,6
Evere	6,0	4,2	4,7	5,4
Forest	6,7	4,7	5,4	6,0
Ganshoren	4,9	3,4	4,0	4,4
Ixelles	7,8	5,4	6,2	6,9
Jette	6,4	4,4	5,1	5,6
Koekelberg	4,9	3,4	4,0	4,4
Molenbeek-Saint-Jean	8,2	5,6	6,5	7,3
Saint-Gilles	6,4	4,4	5,1	5,6
Saint-Josse-Ten-Noode	5,3	3,6	4,2	4,7
Schaerbeek	4,3	3,0	3,4	3,8
Uccle	3,7	2,6	3,0	3,4
Watermael-Boitsfort	4,9	3,4	4,0	4,4
Woluwe-Saint-Lambert	6,7	4,7	5,4	6,0
Woluwe-Saint-Pierre	6,0	4,2	4,7	5,4
TOTAL	5,8	4,0	4,6	5,1

28Table 7 illustrates the difficulties some groups would experience by representing for each municipality the number of lists with less than 3 elected representatives, in the current circumstances and according to the three scenarios. In each of them the number of groups with only one or two elected representatives would increase. In the 30 %scenario that number would increase from 23 to 29, knowing that in addition, a number of lists that only had a single elected representative in 2012 would lose that representative too and disappear from the municipal council. There is no doubt that for some political groups, the decrease in the number of municipal councillors would lead to great difficulties in fulfilling their duties.

Table 7. Number of lists with less than 3 elected representatives, 2012 and projections

Municipality	Number of lists with less than 3 elected representatives in 2012	- 30 %	- 20 %	- 10 %
Anderlecht	3	2	2	2
Auderghem	3	2	2	2
Berchem-Sainte-Agathe	3	4	4	3
Brussels	1	2	1	1
Etterbeek	0	1	1	0
Evere	0	2	1	1
Forest	0	2	1	1
Ganshoren	2	1	2	2
Ixelles	0	1	0	0
Jette	3	2	2	2
Koekelberg	0	1	1	1
Molenbeek-Saint-Jean	3	1	3	3
Saint-Gilles	1	1	1	1
Saint-Josse-Ten-Noode	0	1	1	0
Schaerbeek	0	0	0	0
Uccle	0	1	1	1
Watermael-Boitsfort	1	1	1	1
Woluwe-Saint-Lambert	1	3	2	1
Woluwe-Saint-Pierre	2	1	1	2
TOTAL	23	29	27	24

30In this context, we must also point out that sometimes the recruitment of political staff is very difficult at the local level [Devos, Reynaert and Verlet, 2008]. The lack of time to properly fulfill a mandate was mentioned by various group leaders as being one of the major obstacles to convince people to be a candidate for the elections. If a reduction in the number of councillors can slightly mitigate the problem of recruitment by itself, a concentration of the political work will certainly result in a greater professionalization of the local political mandates, and will certainly also change the profile of the candidates on the lists for the municipal elections [Delwit 2006]. A larger number of councillors, however, can also allow a wider range of opinions, backgrounds and neighbourhoods to be represented in the municipal councils. In conversations we had with 30 local representatives in 2018 (especially Mayors and aldermen) this argument of diversity was very often used to argue against a reduction of the number of councillors.

2.3. The governing majorities

31Another consequence of the reduction of the number of mandates that must be investigated, is that of the formation of governing majorities. Two questions arise in this respect. The first is whether the majority that was formed in 2012 could also be formed after a reduction in the number of councillors. And – secondly – would there be more parties needed to form a majority? Both elements are analysed in the simulations in the tables 8 and 9.

32Table 8 lists for each municipality the percentage of seats controlled by the governing majority that was formed in 2012, and the percentage of seats that they would have after the linear reduction of the number of councillors with 10, 20 or 30 %. In general, a decrease in the number of councillors strengthens the governing majority. Yet the shifts are not very important. A 30 % reduction in the size of the councils leads to an increase of only 1 percentage point. And a reduction in the number of seats with 10 or 20 % produces a shift of more or less one percentage point. In some municipalities, we also see a very slight drop in the size of the coalition. That happens in those cases where the lists that govern the municipality would be most affected by the reduction of the council. That would be the case in Berchem-Sainte-Agathe, Etterbeek, Ixelles, Saint-Gilles, Saint-Josse-Ten-Noode, Schaerbeek and Watermael-Boitsfort. In none of the 19 municipalities the majority of 2012 would be lost by reducing the number of representatives.

Table 8. Percentage of seats controlled by the majorities in the municipalities, in 2012 and projections

Municipality	% of seats in the majority in 2012	- 30 %	- 20 %	- 10 %
Anderlecht	74,5	78,8	76,3	76,22
Auderghem	74,2	77,3	76,0	75,0
Berchem-Sainte-Agathe	59,3	57,9	63,6	62,5
Brussels	57,1	58,8	56,4	56,8
Etterbeek	80,0	79,2	82,1	77,4
Evere	60,6	60,9	61,5	60,0
Forest	64,9	65,4	66,7	63,6
Ganshoren	63,0	68,4	68,2	62,5
Ixelles	53,5	53,3	52,9	52,6
Jette	60,0	62,5	60,7	61,3
Koekelberg	59,3	63,2	63,6	62,5
Molenbeek-Saint-Jean	55,6	58,1	52,8	52,5
Saint-Gilles	71,4	70,8	67,9	71,0
Saint-Josse-Ten-Noode	72,4	70,0	73,9	69,2
Schaerbeek	61,7	60,6	60,5	61,9
Uccle	70,7	72,4	72,7	73,0
Watermael-Boitsfort	55,6	52,6	59,1	54,2
Woluwe-Saint-Lambert	73,0	73,1	73,3	72,7
Woluwe-Saint-Pierre	54,5	56,5	57,7	53,3
TOTAL	64,3	65,3	65,6	64,1

33In table 9, that same weak impact of a decrease in the number of representatives on the formation and the stability of the governing majorities is also evident. In that table we look at the number of parties that are at least required to form a majority (to form a so-called “minimal winning coalition”) in each municipality. That number is almost nowhere influenced by our scenarios, except in one municipality where in the scenario of the 30 % drop, the largest list (the Mayor’s list) would have acquired an absolute majority of the seats.

Table 9. Number of parties necessary to form a minimal winning coalition, in 2012 and projections

Municipality	Minimum number of parties to form a coalition	- 30 %	- 20 %	- 10 %
Anderlecht	2	2	2	2
Auderghem	1	1	1	1
Berchem-Sainte-Agathe	2	2	2	2
Brussels	2	2	2	2
Etterbeek	2	2	2	2
Evere	2	1	2	2
Forest	2	2	2	2
Ganshoren	2	2	2	2
Ixelles	2	2	2	2
Jette	2	2	2	2
Koekelberg	1	1	1	1
Molenbeek-Saint-Jean	2	2	2	2
Saint-Gilles	1	1	1	1
Saint-Josse-Ten-Noode	1	1	1	1
Schaerbeek	2	2	2	2
Uccle	1	1	1	1
Watermael-Boitsfort	2	2	2	2
Woluwe-Saint-Lambert	1	1	1	1
Woluwe-Saint-Pierre	2	2	2	2
TOTAL	1,7	1,6	1,7	1,7

2.4. The Dutch speaking elected representatives

34Any drop in the number of municipal councillors would not only have an impact on the distribution of seats between the lists. It would also have effects on the distribution of seats within the lists. We focus in particular on the effects of a reduction in the number of councillors on the proportion of elected female candidates and on the opportunities for the Dutch speaking candidates. The projections we make are based on the results of the elections in 2012, not on further developments within the lists and in the councils. Again we do not present a neat and complete picture of all possible consequences of a drop in the number of representatives. For example, in the Brussels context it is very relevant to know the extent to which the composition of the lists may have an impact on the voting behaviour of voters with a migration background and for the chances of candidates with a migration background to be elected [Jacobs, 2010]. That is not independent of the extent to which non-Belgian voters register for the local elections [Nikolic, 2017]. And the exact identification of candidates with a migration background is also very difficult [Jacobs, 2009]. For the female and the Dutch-speaking elected representatives on the other hand, it is rather easy to simulate the effects of a reduction in the number of representatives and aldermen.

36That general trend hides large differences between the municipalities. In those municipalities, where the share of the Dutch speaking elected representatives was already limited in 2012, a reduction in the number of mandates can really have a big impact. For example, a 30 % reduction would result in the complete disappearance of Dutch speaking representatives in three municipalities (Auderghem, Forest and Ixelles). They would lose the possibility of adding an additional Dutch speaking alderman to reflect the bilingual character of the BCR. In addition, there would be only one Dutch speaking representative in six municipalities (Saint-Gilles, Saint-Josse-Ten-Noode, Uccle, Watermael-Boitsfort, Woluwe-Saint-Lambert and Woluwe-Saint-Pierre). If that elected representative is not a member of the majority, those municipalities would not appoint any additional Dutch speaking alderman.

Table 10. Number and proportion of the Dutch speaking elected representatives, 2012 and projections

Municipality	Numbers and proportion in 2012		- 30 %		- 20 %		- 10 %
	N	%	N	%	N	%	N	%
Anderlecht	11	23,4	7	21,2	8	21,1	8	19,1
Auderghem	1	3,2	0	0,0	0	0,00	0	0,00
Berchem-Sainte-Agathe	5	18,5	3	15,8	4	18,2	5	20,8
Brussels	10	20,4	8	23,5	10	25,6	10	22,7
Etterbeek	3	8,6	3	12,5	3	10,7	3	9,7
Evere	3	9,1	3	13,0	3	11,5	3	10,0
Forest	0	0,00	0	0,0	0	0,00	0	0,0
Ganshoren	4	14,8	3	15,8	4	18,2	4	16,7
Ixelles	1	2,3	0	0,0	0	0,00	1	2,6
Jette	6	17,1	3	12,5	5	17,9	5	16,1
Koekelberg	5	18,5	3	15,8	4	18,2	4	16,7
Molenbeek-Saint-Jean	5	11,1	2	6,4	3	8,3	5	12,5
Saint-Gilles	3	8,6	1	4,2	2	7,1	2	6,5
Saint-Josse-Ten-Noode	2	6,9	1	5,0	2	8,7	2	7,7
Schaerbeek	3	6,4	2	6,1	2	5,3	2	4,8
Uccle	2	4,9	1	3,4	1	3,0	2	5,4
Watermael-Boitsfort	2	7,4	1	5,3	1	4,6	1	4,2
Woluwe-Saint-Lambert	2	5,4	1	3,8	2	6,7	2	6,1
Woluwe-Saint-Pierre	1	3,0	1	4,3	1	3,9	1	3,3
TOTAL	69	10,1	43	9,0	55	10,0	60	9,8

37This very pernicious effect of a reduction in the number of councillors on the number of Dutch-speaking elected representatives, especially in municipalities where there already are very few of them, is caused by two mechanisms. As already indicated, with a reduction in the number of mandates, the lists that had only one single elected representative in 2012 – in different municipalities those are the monolingual Dutch-speaking lists (N-VA, Vlaams Belang, SAMEN) – are especially vulnerable. On the other hand, the Dutch-speaking elected representatives on bilingual lists often are those who obtain their seat with the lowest number of preference votes. When the number of mandates is reduced, they are the ones who will lose their seat.

2.5. Gender balances

38Electoral systems – and the number of seats available are part of it – are never gender neutral [Kittilson and Schwindt-Bayer, 2012]. However, Table 11 shows that a linear reduction in the number of representatives would not have a negative effect on the number of female elected representatives. Even more so, the projections show that in the three scenarios the number of female elected representatives would slightly rise in the municipal councils of the BCR. In 2012, there were 38,83 % female representatives. That would have been 41,63 % if there were 30 % fewer seats to distribute. It would be 41,27 % in case of a 20 % drop and 40,69 % in case of a 10 % drop. This positive effect is present in all municipalities, except in Berchem-Sainte-Agathe and – for some scenarios – in Brussels-City, Etterbeek and Jette. Finally, we note that a decrease in the number of representatives by 30 % would have led to a female majority in four municipal councils (Auderghem, Ganshoren, Ixelles and Woluwe-Saint-Lambert), whereas this is now only the case in Auderghem and Ixelles.

39This slightly positive effect of the reduction in the number of mandates on the representation of women is not entirely surprising. For municipal elections, the allocation of the list votes to the candidates usually only benefits the first two candidates on the list. The other candidates will obtain their seat based on their preference votes only. Research has shown that male candidates generally receive more preference votes because more male voters only vote for men than female voters only for women [Marien et al. 2016]. The more mandates there are to distribute, the more that benefit of the male candidates in terms of obtained preference votes translates into elected positions. If, however, the number of mandates is reduced, preference votes play a smaller role, and the disadvantage in preference votes for the female candidates does not exist.

Table 11. Number and proportion of elected women in the municipal council, 2012 and projections

Municipality	Elected in 2012		- 30 %		- 20 %		- 10 %
	N	%	N	%	N	%	N	%
Anderlecht	17	36,2	12	36,4	17	44,7	17	40,5
Auderghem	18	58,1	14	63,6	16	64,0	17	60,7
Berchem-Sainte-Agathe	7	26,0	4	21,0	5	22,7	6	25,0
Brussels	18	36,7	12	35,3	14	35,9	17	38,6
Etterbeek	14	40,0	9	37,5	11	39,3	13	41,9
Evere	12	36,4	9	39,1	10	38,5	12	40,0
Forest	14	37,8	12	46,1	13	43,3	13	39,4
Ganshoren	13	48,1	10	52,6	11	50,0	13	54,2
Ixelles	24	55,8	18	60,0	19	55,9	21	55,3
Jette	14	40,0	9	37,5	12	42,9	12	38,7
Koekelberg	9	33,3	8	42,1	9	40,9	9	37,5
Molenbeek-Saint-Jean	11	24,4	8	25,8	8	22,2	10	25,0
Saint-Gilles	13	37,1	10	41, 7	11	39,3	12	38,7
Saint-Josse-Ten-Noode	8	27,6	6	30, 0	7	30,4	7	26,9
Schaerbeek	16	34,0	12	36,4	14	36,8	16	38,1
Uccle	20	48,8	14	48,3	16	48,5	18	48,6
Watermael-Boitsfort	10	37,0	7	36,8	8	36,4	10	41, 7
Woluwe-Saint-Lambert	17	45,9	15	57,7	16	53,3	16	48,5
Woluwe-Saint-Pierre	11	33,3	10	43,5	10	38,5	10	33,3
TOTAL	266	38,8	199	41,6	227	41,3	249	40,7

3. Smaller colleges of Mayor and aldermen

40For the colleges of aldermen we have tested the same three scenarios: a reduction by 10, 20 or 30 % of the number of aldermen. We have done so without taking into account the additional Dutch aldermen (see above) for they are the result of agreements within the majority. We have also applied the three scenarios linearly in each municipality. Table 12 shows the number of aldermen per municipality and for each of the scenarios.

Table 12. Size of the council of aldermen in each of the three scenarios

Municipality	2012-2018	- 30 %	- 20 %	- 10 %
Anderlecht	9	6	7	8
Auderghem	7	5	6	6
Berchem-Sainte-Agathe	6	4	5	5
Brussels	9	6	7	8
Etterbeek	7	5	6	6
Evere	7	5	6	6
Forest	8	6	6	7
Ganshoren	6	4	5	5
Ixelles	8	6	6	7
Jette	7	5	6	6
Koekelberg	6	4	5	5
Molenbeek-Saint-Jean	8	6	6	7
Saint-Gilles	7	5	6	6
Saint-Josse-Ten-Noode	6	4	5	5
Schaerbeek	9	6	7	8
Uccle	8	6	6	7
Watermael-Boitsfort	6	4	5	5
Woluwe-Saint-Lambert	8	6	6	7
Woluwe-Saint-Pierre	7	5	6	6
TOTAL	139	97	111	125

41The most drastic scenario of 30 % would reduce the number of aldermen in the BCR by 42 units (from 139 to 97). A 20 % linear reduction would reduce the number of aldermen by 28 and a 10 % reduction would result in a drop of 14 aldermen. In case of a 30 % reduction, the smallest councils in the BCR would count 4 aldermen, or the number of aldermen stated in the NLC for municipalities with 5 000 to 9 999 inhabitants.

42The reduction in the number of aldermen leads to the same considerations as the reduction in the number of councillors. Should the colleges all have an odd number of aldermen? Should there be a minimum per municipality? Can an upper limit be set for the whole BCR, after which the fixed number of aldermen for the region is distributed among the municipalities on the basis of the population?

43The question also arises as to how the work will be able to be carried out and distributed if there are fewer aldermen. The current workload of the aldermen varies greatly from one municipality to another, and also between the aldermen within the municipalities. It is therefore very difficult to assess exactly what the impact would be. However, the frequency of the accumulation of local mandates and regional, Federal and European mandates (see above) suggests that there is room to increase the workload for the representatives.

45As with the reduction in the number of municipal councillors, a reduction in the number of aldermen would also have a clear and negative impact on the number of Dutch speakers in the colleges. In 2012, there were only five municipalities that had a Dutch speaking college member, without the application of the mechanism of the additional aldermen (Berchem-Sainte-Agathe, Brussels-City, Etterbeek, Jette, and Molenbeek-Saint-Jean). With fewer Dutch speaking elected representatives, the chances of one of them being able to become an alderman clearly decrease. There is a risk that the access for Dutch speakers to the college of aldermen will only be done through the mechanism of the additional aldermen, and in a smaller number – a small dozen – of municipalities than is currently the case. Since a decrease in the number of municipal councillors may also result in the fact that there are few or no Dutch speaking elected representatives in different municipalities, there is of course an impact on the possible participation in the majority and thus on the composition of the colleges of aldermen.

46For the gender balances too, a reduction in the colleges of aldermen could have a strong and negative impact. With seven municipalities that have only one or two female aldermen, a reduction in the number of aldermen could work against them.

Conclusions

48This relatively high number of representatives – even when compared to other (capital) cities – often leads to suggestions to reduce the number of positions. The Brussels Minister-President Vervoort too, has already said several times that he wants to move in that direction. An ordinance dated 25 January 2018 stipulates that the number of aldermen in 2018 will remain the same as in 2012. That is not a reduction, but the non-implementation of a possible increase in the number of aldermen.

49In this contribution, we examined the impact of a possible reduction in the number of councillors and aldermen. We analysed three scenarios: a reduction of 10, 20 and 30 %. We can conclude that a decrease in the number of local representatives would not be a politically neutral process. First of all, we note that the impact on the smaller groups in the municipal councils would be very high. Those would become even smaller, thus distributing the workload over a smaller number of councillors. In addition, it is also very clear that the decline in the number of municipal councillors would especially affect the Dutch speaking elected representatives. Their number would drop in all scenarios, and in some municipalities would even be reduced to one or zero. There are only a limited number of lists that would disappear from the municipal councils in the case of a decline in the number of councillors, but these are almost all Dutch speaking lists, and in particular the unilingual Dutch speaking lists. The Dutch speakers would also be affected by the composition of smaller colleges of aldermen. In municipalities with only one Dutch speaking elected representative, the probability of being included in the college of aldermen is less likely, and where there are no more Dutch speaking elected representatives, the mechanism of the additional alderman can no longer be used.

50There would not be any impact of the drop in the number of councillors on gender balances. But for the colleges of aldermen we do expect visible consequences. Women are already strongly under-represented in the colleges – seven municipalities have only one or two female aldermen – and they would have even less chance to be appointed in smaller colleges.

51A reduction of the number of local representatives in the Brussels-Capital Region is certainly legally possible. However, its political impact is not negligible. In the debates that will certainly continue to follow on the status of the municipalities, their number, their relationship to the regional level and the necessary or unnecessary complexity of the institutions, the knowledge that a relatively simple change such as the linear reduction of the number of councillors or aldermen can have serious consequences, is a very relevant insight.