1The article examines the residential distribution of elected representatives throughout the Brussels-Capital Region, excluding the municipal level. We have therefore taken into consideration the elected members of the regional parliament, the elected members of the federal parliament and the European parliament, and the regional and federal ministers and secretaries of state living in Brussels, the vast majority of whom are elected representatives who are prevented from sitting. The frame of reference for the assessment of the residential distribution of political staff is therefore the Region as a whole, unlike studies concerning elected representatives or candidates in municipal elections, which therefore had to use the various municipalities separately as frames of reference [de Borman et al., 2001; Van Hamme and Marissal, 2008; de Maesschalck, 2009]. For this study, we chose to work within the framework of urban structures on a regional scale. However, we also took into consideration the mayors of the 19 Brussels municipalities, as their small number and political weight at regional level justify the use of this scale of reference with respect to their place of residence. Two studies had already been conducted previously in Brussels on a regional scale [de Maesschalk, 2010; Vandermotten, 2014].
- 1 For the sake of simplicity, we shall refer to 2010 and 2020 for the two situations studied.
2The diachronic analysis conducted here compares two electoral cycles. On the one hand, it considers the situation in 2014, on the eve of the federal and regional elections. It therefore concerns candidates elected for the most part in the 2012 municipal elections (for mayors), the 2009 regional and European elections and the 2010 federal elections, unless they were subsequently replaced by deputy members. On the other hand, it considers the situation at the beginning of 2021, which is the result of the 2018 municipal elections (for mayors) and the 2019 regional, federal and European elections.1 The two data sets are therefore separated by an intermediate, unexamined election cycle.
3The article examines two issues, which are dealt with separately.
4The first is that of the residential pattern of elected representatives. It is underpinned by the idea of the equitable representation of different types of neighbourhood among elected representatives, and the political significance of the under-representation of disadvantaged neighbourhoods in their places of residence or, conversely, the over-representation of affluent neighbourhoods. The implicit hypotheses are therefore that elected representatives are to some extent representative of the social structure of the neighbourhoods they live in, that they may prefer to relay the concerns of the categories of population living in these neighbourhoods, especially on the scale of an urban region – which would probably be less true on a larger scale – and that the residential choices of elected representatives also contribute to their own representation of the city and the urban project.
- 2 For example, if the far right scored high in neighbourhoods where there is a high percentage of Bel (...)
5However, it is clear that the results should be interpreted with caution: there is no way to deduce with certainty that elected representatives benefit preferentially from the vote of residents of their neighbourhood, nor that it represents their electoral base. It has also been shown that on average, and especially in disadvantaged neighbourhoods, elected representatives belong to the more well-to-do social classes than the population of their neighbourhood as a whole, and also than their average voters [Jacobs, 2006; Van Hamme and Marissal, 2008; de Maesschalck, 2010]. The results should therefore be interpreted with caution as regards the classic risks of ecological fallacy, i.e. the interpretation of a statistical concordance in terms of an explanation2, especially as we are analysing a population whose size is limited. We must also beware of improper semantic shifts, which would lead to the automatic assimilation of the location of types of elected representative and types of voters.
6These precautions apply all the more to the second question in the study, namely that of the parallels between the geography of elected representatives and electoral geography, with the difficulty that the data relating to the latter are at best only available at municipal level and only for elections held at that level.
- 3 That is, in 2010, 89 regional parliamentarians (including 12 mayors), 14 federal parliamentarians ( (...)
7The population studied – whose home addresses or at least the neighbourhood of residence needed to be identified – was made up of 125 elected representatives (or ministers) in 2010 and 133 in 2020.3 The increase reflects a drastic reduction in the number of functions held concurrently by mayors: out of the 19 Brussels mayors, 14 held another elective or ministerial function in 2010, whereas 10 years later only 5 of them did.
8Table 1 and Figure 1 describe the residential pattern of this political staff at municipal and neighbourhood level, as delimited by Monitoring des quartiers [IBSA, accessed on 1 March 2021].
Table 1. Distribution of elected representatives according to municipality of residence and type of neighbourhood (2010 and 2020)
|
Elected representatives
|
Elected representatives (%)
|
Population (%)
|
Ratio % elected rep./% pop.
|
|
2010
|
2020
|
2010
|
2020
|
2010
|
2020
|
2010
|
2020
|
Brussels-Capital Region
|
125
|
133
|
100,0
|
100,0
|
100,0
|
100,0
|
1,00
|
1,00
|
Brussels-Centre, North-east, Louise
|
14
|
6
|
11,2
|
4,5
|
6,7
|
6,8
|
1,68
|
0,66
|
Schaerbeek
|
12
|
15
|
9,6
|
11,3
|
11,1
|
10,9
|
0,86
|
1,04
|
Saint-Josse
|
5
|
4
|
4,0
|
3,0
|
2,4
|
2,2
|
1,65
|
1,34
|
Etterbeek
|
5
|
3
|
4,0
|
2,3
|
4,1
|
4,0
|
0,98
|
0,57
|
Ixelles
|
16
|
9
|
12,8
|
6,8
|
7,4
|
7,2
|
1,74
|
0,94
|
Saint-Gilles
|
5
|
6
|
4,0
|
4,5
|
4,3
|
4,1
|
0,93
|
1,11
|
Forest
|
6
|
13
|
4,8
|
9,8
|
4,6
|
4,7
|
1,04
|
2,10
|
Anderlecht
|
8
|
9
|
6,4
|
6,8
|
9,6
|
9,9
|
0,67
|
0,68
|
Molenbeek
|
7
|
10
|
5,6
|
7,5
|
8,1
|
8,0
|
0,69
|
0,94
|
Koekelberg
|
1
|
6
|
0,8
|
4,5
|
1,8
|
1,8
|
0,44
|
2,50
|
Brussels-Laeken
|
5
|
9
|
4,0
|
6,8
|
5,8
|
6,0
|
0,69
|
1,13
|
Berchem Sainte-Agathe
|
2
|
2
|
1,6
|
1,5
|
2,0
|
2,1
|
0,79
|
0,71
|
Ganshoren
|
4
|
3
|
3,2
|
2,3
|
2,1
|
2,1
|
1,54
|
1,08
|
Jette
|
6
|
6
|
4,8
|
4,5
|
4,3
|
4,3
|
1,12
|
1,04
|
Brussels-Neder-over-Hembeek
|
2
|
0
|
1,6
|
0,0
|
1,6
|
1,8
|
1,02
|
0,00
|
Brussels-Haren
|
0
|
0
|
0,0
|
0,0
|
0,4
|
0,5
|
0,00
|
0,00
|
Evere
|
3
|
5
|
2,4
|
3,8
|
3,3
|
3,5
|
0,73
|
1,07
|
Woluwé Saint-Lambert
|
3
|
1
|
2,4
|
0,8
|
4,7
|
4,8
|
0,52
|
0,16
|
Woluwé Saint-Pierre
|
7
|
8
|
5,6
|
6,0
|
3,6
|
3,5
|
1,56
|
1,73
|
Auderghem
|
1
|
4
|
0,8
|
3,0
|
2,8
|
2,8
|
0,28
|
1,06
|
Watermael-Boitsfort
|
5
|
5
|
4,0
|
3,8
|
2,2
|
2,1
|
1,80
|
1,80
|
Uccle
|
8
|
9
|
6,4
|
6,8
|
7,1
|
6,9
|
0,90
|
0,98
|
Neighbourhoods in the first income quintile
|
23
|
32
|
18,4
|
24,1
|
26,9
|
26,4
|
0,68
|
0,91
|
Neighbourhoods in the second income quintile
|
33
|
27
|
26,4
|
20,3
|
20,7
|
21,0
|
1,28
|
0,97
|
Neighbourhoods in the third income quintile
|
22
|
28
|
17,6
|
21,1
|
20,4
|
20,9
|
0,86
|
1,01
|
Neighbourhoods in the fourth income quintile
|
19
|
19
|
15,2
|
14,3
|
18,5
|
18,8
|
0,82
|
0,76
|
Neighbourhoods in the fifth income quintile
|
28
|
27
|
22,4
|
20,3
|
13,5
|
12,9
|
1,66
|
1,57
|
Pentagon area
|
11
|
3
|
8,8
|
2,3
|
4,5
|
4,5
|
1,96
|
0,50
|
Poor area
|
17
|
28
|
13,6
|
21,1
|
19,8
|
19,2
|
0,69
|
1,10
|
Eastern part of the inner ring
|
28
|
23
|
22,4
|
17,3
|
18,3
|
17,9
|
1,22
|
0,97
|
Eastern part of the outer ring Q1-Q3
|
11
|
15
|
8,8
|
11,3
|
10,2
|
10,7
|
0,86
|
1,05
|
Eastern part of the outer ring Q4-Q5
|
28
|
31
|
22,4
|
23,3
|
21,7
|
21,4
|
1,03
|
1,09
|
Western part of the outer ring Q1-Q3
|
24
|
26
|
19,2
|
19,5
|
21,3
|
21,9
|
0,90
|
0,89
|
Western part of the outer ring Q4-Q5
|
6
|
7
|
4,8
|
5,3
|
4,3
|
4,4
|
1,12
|
1,20
|
Neighbourhoods undergoing gentrification
|
46
|
43
|
36,8
|
32,3
|
35,1
|
34,3
|
1,05
|
0,94
|
(a) Neighbourhoods categorised according to their 2015 average income divided into five quintiles from Q1 (lowest income neighbourhoods) to Q5 (highest income neighbourhoods).
(b) The neighbourhoods of the Pentagon, the poor area and the eastern part of the inner ring which were in the first two income quintiles in 2005 and whose relative average income/inhabitant increased between 2005 and 2015.
Figure 1. Residential pattern of elected representatives in 2010 and 2020
Cartography: Pierre Marissal
The symbols are located in the neighbourhood of residence of the elected representatives, with a circle for the French-speaking representatives and a square for the Dutch-speaking representatives. The colours linked with political groups are used. We have distinguished between: the far left (PTB-PVDA); social democrats, including dissidents Emir Kir and Emin Özdara; the green parties, Ecolo and Groen; the MR and the FDF/DéFI elected representatives, grouped together because of the similar geography and sociology of their voters (whereas those of Open Vld differ significantly); the Social Christians (whose geography and sociology differ greatly for the French- and Dutch-language parties); the N-VA and the far right.
9In terms of differences between municipalities, it would be unwise to draw conclusions regarding the residential geography of elected representatives: the absolute numbers are small, and most municipalities are mixed. It is therefore necessary to start from the neighbourhood level and group them into coherent sociological and geographical units.
10The neighbourhoods were categorised according to two methods (Figure 2): on the one hand, a distribution in five quintiles of average income/inhabitant (Q1 to Q5); on the other hand, a geography distinguishing between the Pentagon area, the poor area outside the Pentagon (which includes the entire western part of the inner ring), the eastern part of the inner ring outside the poor area, and the outer ring, separating the neighbourhoods located east and west of the Senne and distinguishing between low- and medium-income neighbourhoods (quintiles 1 to 3, i.e. Q1-Q3) and the most affluent ones (quintiles 4 and 5, i.e. Q4-Q5) (Table 2).
Table 2. Relative average income per inhabitant in the different types of neighbourhood
Brussels-Capital = 100
|
Share of the population (2015)
|
Relative average income/inhabitant
|
2005
|
2015
|
Pentagon
|
4,5
|
78,9
|
82,3
|
Poor area
|
19,9
|
58,5
|
62,2
|
Neighbourhoods in the eastern part of the inner ring
|
18,1
|
58,5
|
62,2
|
Neighbourhoods in the eastern part of the outer ring Q1, Q2, Q3
|
10,3
|
100,9
|
95,6
|
Q4, Q5
|
21,2
|
129,8
|
136,4
|
Neighbourhoods in the western part of the outer ring Q1, Q2, Q3
|
21,7
|
103,9
|
94,9
|
Q4, Q5
|
4,3
|
133,5
|
124,3
|
Figure 2. Categories of neighbourhoods in Brussels
Cartography: Pablo Medina Lockhart.
11This distribution highlights the east-west social dichotomy in Brussels. The inner ring is home to 38 % of the population of Brussels, with about half in the poor area in the north and west, and the other half in the wealthier, historically middle-class neighbourhoods in the east. In the outer ring, 32 % of the Brussels population lives in the east and 26 % in the west, but 67 % of the population of the eastern part of the outer ring lives in neighbourhoods belonging to income quintiles 4 and 5, whereas this percentage is only 17 % in the west (Table 2).
- 4 It should be borne in mind, however, that income statistics must be used with caution, especially w (...)
12Finally, the neighbourhoods which are undergoing gentrification were isolated, based on a very rough definition of gentrification, namely the neighbourhoods in the Pentagon and inner ring which were in the first two income quintiles in 2005 and whose relative average income/inhabitant increased between 2005 and 20154 (Table 2).
13The first research question is therefore whether the places of residence of the elected representatives are evenly distributed among the different neighbourhood categories in the city.
- 5 At the level of neighbourhoods classified according to income quintile, the gap between the distrib (...)
14There is no exaggerated discrepancy between the distribution of the residences of elected representatives and that of the general population, with the exception of a very clear over-representation of the wealthiest neighbourhoods (Table 1). In addition, the residential patterns of elected representatives and the population as a whole tend to converge between 2010 and 2020, as the under-representation of the poorest neighbourhoods is reduced.5
15This improvement in the alignment between the distribution profile of elected representatives and that of the population as a whole as a result of better representation of elected representatives in less privileged neighbourhoods may be due to various interrelated factors, which we shall now examine:
-
changes in electoral results, benefiting the parties whose elected representatives are present in these neighbourhoods;
-
- 6 Contrary to popular belief, the turnover of elected representatives from one election to the next i (...)
the turnover of political staff, partly linked to these developments (Table 3)6 ;
-
- 7 In the study, two groups are identified in particular: on the one hand, elected representatives of (...)
in particular, a growing number of elected representatives from an immigrant background7;
-
the spread of gentrification phenomena to all disadvantaged neighbourhoods, manifested in an increase in relative income (Table 4) and associated with the election and increase in the number of new younger candidates, such as those from the green parties.
Table 3. Changes in political representation and turnover rate of elected representatives
|
Percentage of all elected representatives in 2010
|
Percentage of all elected representatives in 2020
|
Percentage of 2020 elected representatives already elected in 2010
|
PTB/PVDA
|
-
|
10,5
|
0,0
|
Socialists (including SP.A and dissidents)
|
29,6 (b)
|
24,8 (b) ↓
|
54,5
|
Green parties (Ecolo and Groen)
|
20,8 (c)
|
23,3 (c) ↑
|
34,4
|
MR
|
16,0 (d)
|
14,3 (d) ↓
|
36,9
|
FDF/DéFI
|
10,4 (e)
|
10,5 (e) =
|
35,7
|
CdH
|
11,2 (a)
|
7,5 (a) ↓
|
60,0°
|
Flemish parties (including SP.A and Groen)
|
18,4
|
14,3 (f) ↓
|
35,0
|
Total
|
100,0 (n = 125)
|
100,0 (n = 133)
|
38,3
|
(a) Including Bertin Mampaka, who has since joined MR.
(b) Including Emir Kir and Emin Özkara, who have since become independent.
(c) Including Marie Nagy who has since joined DéFI.
(d) Excluding Bertin Mampaka.
(e) Excluding Marie Nagy.
(f) Excluding the two Flemish elected representatives from the PTB/PVDA.
The stability of the elected representatives is strongest in the two political families which lost the most between 2010 and 2020, namely the socialists and the CdH, which reflects the fact that in a situation of electoral decline they were less able to get new candidates elected and that the most established elected representatives were more likely to remain. Thus, for the CdH, the electoral decline did not prevent some mayors from remaining in office, and at the same time staff turnover seems to be low. For the socialists, the continued existence of “historical” elected representatives involves those from an immigrant background as well as those who are not. In contrast, the PTB/PVDA had no elected representatives in 2010. The other political groups are in an intermediate situation (approximately 2/3 of the elected representatives from 2020 were not in office in 2010).
Table 4 Distribution of elected representatives according to type of neighbourhood, political family and background
(a) The neighbourhoods of the Pentagon, the poor area and the eastern part of the inner ring which were in the first two income quintiles in 2005 and whose relative average income/inhabitant increased between 2005 and 2015.
16Contrary to what one might have expected, the increased representation of the poorest neighbourhoods (Q1) is not so much due to the appearance of far-left elected representatives (only 2 out of the 32 elected representatives or out of the 14 far-left elected representatives in Brussels were living in these neighbourhoods in 2020), but rather to a very clear increase in the number of green elected representatives, especially Flemish: the attractiveness of the canal area among this type of elected representative is remarkable. With 41 % of elected representatives from neighbourhoods in the first quintile in 2020 compared to 22 % in 2010, the green parties have made the area their “residential stronghold”. On the other hand, the number of socialist elected representatives is decreasing in these neighbourhoods where they used to dominate: if we consider the neighbourhoods in the first two quintiles, they have gone from 23 to 16 elected representatives, while the number of green elected representatives has increased from 12 to 19. Therefore, from 2010 to 2020, while the numbers of socialist and green elected representatives have remained fairly stable in middle-class and affluent neighbourhoods (Q3 to Q5), their predominance is reversed in disadvantaged neighbourhoods.
- 8 It should be noted that although the loyalty of the electorate of Maghrebi and Turkish origin to th (...)
17However, it would be wrong to link this development to a possible change in the electoral behaviour of the local lower classes from an immigrant background, who traditionally have not voted for the green parties [Martiniello and Hily, 1998; Rea et al., 2010; Teney et al. 2010]. Table 5 shows that there has not been an appreciable decrease in the very significant number of socialist elected representatives from a Maghrebi or Turkish immigrant background in these neighbourhoods, and that there are still very few green elected representatives from this immigrant background despite the openness of the green parties to ethnic and immigration issues, despite the fact that the green party was the first party in Brussels to have an elected representative of Moroccan origin, and despite the fact that it has been co-chaired by a Brussels woman of Maghrebi origin twice: only 4 out of 31 green elected representatives from 2020, i.e. 13 %, are of Maghrebi or Turkish origin, whereas 19 out of 45 socialist and far-left elected representatives, i.e. 42 %, are of Maghrebi or Turkish origin.8
Table 5. Number of elected representatives of Maghrebi and Turkish origin and of sub-Saharan origin, according to type of neighbourhood and political affiliation
|
2010
|
2020
|
Social.
|
Green
|
Other
|
Total
|
Social. and PTB
|
Green
|
Other
|
Total
|
Total
|
Elected representatives of Maghrebi or Turkish origin
|
Q 1 and Q 2
|
12
|
2
|
3
|
17
|
11
|
3
|
2
|
16
|
Q 3, 4 and 5 east
|
2
|
2
|
-
|
4
|
3
|
-
|
1
|
4
|
Q 3, 4 and 5 west
|
1
|
-
|
-
|
1
|
5
|
1
|
-
|
6
|
Elected representatives of Sub-Saharan origin
|
Q 1 and Q 2
|
-
|
-
|
1
|
1
|
-
|
-
|
1
|
1
|
Q 3, 4 and 5 east
|
1
|
-
|
1
|
2
|
-
|
-
|
-
|
-
|
Q 3, 4 and 5 west
|
-
|
-
|
3
|
3
|
-
|
1
|
3
|
4
|
Total
|
16
|
4
|
8
|
28
|
19
|
5
|
7
|
31
|
The surnames of the elected representatives have been used to determine which category they belong to. While this is a good proxy variable for the origin of elected representatives, we would like to point out that its use has its limitations. A person's origins can be much more complex than what his or her surname might indicate. For example, this explains why Barbara Trachte, who has partly sub-Saharan origins, was not included in this table.
18We might therefore conclude that the growth in the number of green elected representatives residing in disadvantaged neighbourhoods is the result of the residential choices of pioneers in the gentrification process, who are not very representative of the dominant political behaviour of local populations of immigrant origin. Conversely, a proportion of the socialist political elite from an immigrant background tends to move to more affluent areas, especially in the western part of the Region.
19In these disadvantaged neighbourhoods (Q1 and Q2), elected representatives from centre and right-wing parties are increasingly in the minority (41 % of all elected representatives residing in these neighbourhoods in 2010, and 27 % in 2020), although their numbers are somewhat supported by a Flemish presence, which is nevertheless also declining overall.
20Left-wing elected representatives (far left, social democrats and green) are also in the majority among those living in intermediate neighbourhoods (Q3 and Q4), but less so than in the disadvantaged neighbourhoods: 60 % of elected representatives in Q3 and Q4 neighbourhoods are from left-wing parties (64 % in Q3 neighbourhoods, 53 % in Q4 neighbourhoods), compared to 70 % in Q1 and Q2 neighbourhoods.
21Within the group of left-wing elected representatives, the weight of those from the socialist party has increased in the intermediate neighbourhoods compared with those from the green parties, and the latter are better represented in Q3 than in Q4, which once again points to the increase in the number of pioneers in the gentrification process or at least of green elected representatives in search of a dense and central urban neighbourhood.
22But what is less expected is the strong presence of far-left elected representatives in these intermediate neighbourhoods, whose numbers are much higher than in the most disadvantaged neighbourhoods. The preference of PTB-PVDA elected representatives to live in the intermediate neighbourhoods of the eastern part of the inner ring could perhaps be explained by the fact that this first generation of far-left elected representatives does not come mainly from the working classes of immigrant origin and their neighbourhoods of residence, but from a highly politicised trade union and associative sector, with some of them being politically committed academics. Only 3 out of 14 of the PTB-PVDA elected representatives are of Maghrebi or Turkish origin compared with 14 out of 31 elected representatives from the socialist party, which has been able to build up a constituency of immigrant origin at the level of municipal support structures, including among the self-employed and small business owners.
23As far as the elected representatives from the centre and the right are concerned, the (declining number of) Social Christian elected representatives are concentrated increasingly in the intermediate neighbourhoods, in particular those in the western part of the outer ring. This is probably related to the fact that there is a concentration of people of sub-Saharan origin in these neighbourhoods. The profile of elected representatives of sub-Saharan origin is completely different from that of elected representatives of Maghrebi or Turkish origin, as is the general spatial pattern of this population: they live mainly in middle-class neighbourhoods in the outer ring, especially in the west, where the first and only person of African origin was elected mayor of a municipality of Brussels, namely Ganshoren, representing the Social Christian party (Table 5).
24In the neighbourhoods in the fifth quintile – which are over-represented in terms of the residence of elected representatives in relation to the population size – left-wing elected representatives are clearly in the minority this time, with only 29 % of all elected representatives. Among them, the green elected representatives are on a par with the socialists.
25The fairly similar spatial distribution of Liberal and DéFI elected representatives is very unequal: 73 % of the elected representatives of this political family lived in neighbourhoods in the two most affluent quintiles in 2010, and 63 % in 2020 (note that these neighbourhoods represent only 32 % of the population of Brussels). Moreover, their numbers are much higher in the east of the Region than in the west: 26 of the 35 elected representatives from 2020 live in the eastern part of the inner and outer rings, and only 8 in the poor area and the western part of the outer ring. This reflects the historical concentration of the wealthy – and French-speaking – social classes in the eastern suburbs. Within the MR/Open Vld - FDF/DéFI group of elected representatives, the share of the MR tends to increase with the social level of the neighbourhoods: the elected representatives of DéFI and their voters seem to belong more to the middle classes than those of the MR. This more middle-class profile is also seen with the four Flemish liberal elected representatives from Open Vld.
26In these well-to-do neighbourhoods, a significant number of CdH elected representatives still exists only in Woluwé Saint-Pierre. This municipality lies within the area of the historical radial expansion of the Catholic elite in the direction of the suburbs [Vandermotten and Ots-Albitar, 1976; Dobruszkès and Vandermotten, 2006].
27Among the most significant conclusions which emerge from the geography of elected representatives, the following points should be highlighted:
-
the places of residence of elected representatives are increasingly in line with those of the population as a whole, other than the continued over-representation of the most affluent neighbourhoods;
-
the increase in the number of elected representatives who live in the most disadvantaged neighbourhoods is mainly due to the arrival of pioneers in the gentrification process from the green parties, and is not reflected in an increase in the proportion of green elected representatives among those of Maghrebi or Turkish origin;
-
the number of elected representatives from this immigration background is increasing and they are mainly from the social democratic or far-left parties. They live increasingly in intermediate neighbourhoods, although the majority still live in disadvantaged neighbourhoods;
-
the places of residence of the far-left elected representatives suggest that they are political staff from a qualified and politicised working world or politically committed academics who live in intermediate neighbourhoods;
-
the number of elected representatives from the centre and the right increases with the social level of the neighbourhoods. The vast majority of them are MR or FDF/DéFI elected representatives. The more affluent the neighbourhood, the greater the proportion of the former;
-
today, the Social Christian elected representatives are a residual group, which has a significant presence only in two areas: the intermediate neighbourhoods of the western part of the outer ring and, among the affluent neighbourhoods, Woluwé Saint-Pierre;
-
the Dutch-speaking elected representatives from the centre and the right (CD&V and Open Vld) tend to live in the intermediate neighbourhoods of the western part of the outer ring. In general, they have a more working-class profile than their French-speaking counterparts.
28It is difficult to examine in detail the coherency between the residential geography of elected representatives and electoral results, as electoral results are only available at best at municipal level, and even then, only for municipal elections (2012 and 2018) (Figure 3). However, the division of municipalities is heterogeneous and far from being the best suited in terms of understanding social differentiation in the urban space.
Figure 3. Main political trends at municipal level (2018 municipal elections)
Cartography: Pablo Medina Lockhart
29Nevertheless, an exploratory attempt can be made to understand the similarity or lack thereof between the residential distribution of elected representatives and electoral results at municipal level. In order to reduce as much as possible the statistical hazards of small numbers, neighbouring municipalities with similar socio-political profiles have been grouped together: Berchem Sainte-Agathe, Ganshoren and Jette; Molenbeek and Koekelberg; Ixelles and Etterbeek; the two Woluwés; and Auderghem and Watermael-Boitsfort. Similarly, the political families were divided into six groups only: the far left, the socialists, the green parties, the liberals and DéFI, the Social Christians and the NVA and Flemish far right.
- 9 Number of votes per party in each municipality as a percentage of the total number of valid votes i (...)
Table 6. Number of resident elected representatives and expected number based on proportionality with vote distribution in 2012 and 2018 municipal elections9
Residential locations which are more numerous than expected are highlighted in red, and those which are less numerous, in blue.
30The comparison between the number of elected representatives expected based on the breakdown of votes according to political group and group of municipalities and the actual number of elected representatives from the different groups residing in these groups of municipalities must be interpreted with great caution in terms of statistical significance, and does not reveal any major surprises: overall, the geography of the elected representatives reflects that of the electorate (Table 6). Nevertheless, a few revealing differences emerge, which confirm the conclusions already outlined above:
-
the elected representatives from the far left live in more intermediate neighbourhoods than those of a significant proportion of their electorate;
-
the green elected representatives are proportionally better represented than their electorate in the disadvantaged neighbourhoods, and this is becoming more pronounced: thus, the over-representation of green elected representatives in the intermediate neighbourhoods of Ixelles and Etterbeek, in those of the western part of the outer ring, and even in Watermael-Boitsfort and Auderghem, which prevailed in 2012, has disappeared, whereas it has been reinforced in the neighbourhoods along the canal, as well as in Saint-Josse and Saint-Gilles;
-
although very unevenly distributed across the Region, the places of residence of DéFI and liberal elected representatives, especially the latter, best reflect the geography (which is also very unevenly distributed) of their voters.
31This article does not claim that there is a perfect coherence between the distribution of the elected representatives and the voters of the different political families. However, beyond the interest of a geography of elected representatives, enough converging conclusions emerge in order to specify some major features of the socio-political geography of Brussels and its recent developments, which allow us to refine our findings from the analysis of the results of the municipal elections, i.e. the most detailed spatial scale available, which are sometimes difficult to interpret because of the heterogeneity of the municipal areas, or even because of the local “loyalties” of the electorate, in particular to their mayors.
32The PTB/PVDA elected representatives do not live in the poorest neighbourhoods so much as in middle-income areas, reflecting the fact that they come from a highly politicised environment of skilled labour, seasoned trade union leaders and even the academic world.
33Although the most disadvantaged neighbourhoods with a large number of immigrants of Maghrebi or Turkish origin still vote overwhelmingly for the socialists, the residential pattern of socialist elected representatives shows a shift from the central neighbourhoods to more affluent ones, even in the case of elected representatives from an immigrant background. These locations coincide with those of socialist voters “of Belgian origin” linked to the middle classes and the non-market sector. As in other major European cities, the socialist world must therefore deal with two very different electoral segments with conflicting interests, with the risk of a gradual detachment of its upwardly mobile political staff from its most working-class electoral base.
34Increasingly, the green elected representatives fall into the category of pioneers in the gentrification process in the central neighbourhoods, particularly in the canal area, without indicating that this political movement has won over a significant amount of the voters from an immigrant background who live there. However, it should not be overlooked that there is still a significant number of green elected representatives in certain areas of the outer ring (13 of their elected representatives in 2010 and 12 in 2020 live there), and that among the three municipalities where they won the last municipal elections (Ixelles, Watermael-Boitsfort and Forest), the first is located in the eastern part of the inner ring as well as the outer ring, the second is located in the outer ring, and the third is located in the poor area as well as the outer ring. The green elected representatives therefore seem to be divided between those who are part of the gentrification process, who are probably younger and are drawn to the social and cultural aspects of urban life, and others who probably have privileged ties to the non-market sector and are more interested in the preservation of a quiet living environment in the green neighbourhoods of the outer ring.
35The centre and right-wing electorate votes overwhelmingly for the liberals and DéFI. It is mainly concentrated in the intermediate and especially in the affluent neighbourhoods of the outer ring. The residential geography of the elected representatives of these two political families reflects that of their voters, with, however, a tendency towards the middle classes in the case of DéFI, and the wealthiest classes in the case of the liberals. In 2020, 63 % of the elected representatives of this political group were living in neighbourhoods in the two most affluent quintiles, which account for only 32 % of the Brussels population. Moreover, their numbers are much higher in the east of the Region than in the west: 26 of the 35 elected representatives from 2020 live in the eastern part of the inner and outer rings, and only 8 in the poor area and the western part of the outer ring. As with many other aspects of the social geography of Brussels, this reflects the historical concentration of the wealthy – and French-speaking – social classes in the eastern suburbs.
36Spatial inequality in the distribution of CdH elected representatives has increased significantly, reflecting the sharp decline in their numbers and the concentration in specific niches of those remaining. CdH voters and elected representatives are more likely to live in the western part of the outer ring, in particular where they still have power at municipal level (Berchem Sainte-Agathe, Ganshoren and Jette). Elected representatives from a sub-Saharan background represent a significant proportion of CdH voters. Furthermore, the CdH retains a significant weight in the affluent neighbourhoods of Woluwé Saint-Pierre, where it can also rely on the position of mayor.
- 10 This explains why 77 % of Dutch-speaking elected representatives are regional parliamentarians, whi (...)
- 11 Note that the municipalities of Berchem Sainte-Agathe and Ganshoren (as well as Evere) were not inc (...)
37With a total number of seats in the regional parliament which is decreasing despite a guaranteed representation of 17 out of 89 seats,10 and despite the fact that they are very much in the minority everywhere while showing a trend towards a more balanced distribution, there are more Dutch-speaking elected representatives in the west, with a shift towards the poor area (3 elected representatives in 2010, 7 in 2020). To this should be added the elected representatives who live in the Pentagon area (4 in 2010, 1 in 2020). Thus, it seems that there are two Dutch-speaking electorates and two corresponding categories of elected representatives: on the one hand, a historical electorate in the western part of the outer ring, in particular in the former village centres11; on the other hand, an electorate of those who are part of the process of gentrification and the cosmopolitan middle classes drawn to certain parts of the poor area, initially located around Rue Dansaert and then along the canal. In 2020, 3 of the 6 Groen elected representatives were living in this area.