On July 15, Philippe Van Parijs published an excellent piece analysing the Brussels electoral system, exposing its flaws and putting forward constructive proposals for a thorough overhaul. Among the many valid points he makes, however, several elements call for further nuance — especially from a Flemish perspective. That is the aim of this response.
A multi-party system was never the problem
Van Parijs opens by noting that no fewer than 14 parties sit in the Brussels regional parliament, a degree of fragmentation found in neither Flanders (7 parties) nor Wallonia (5 parties). At first glance this seems compelling — but it simply reflects Brussels' bilingual character, in contrast to the linguistically homogeneous Flanders and Wallonia. Indeed, the Belgian federal parliament itself counts 12 parties, and the Dutch parliament has 15, despite being entirely monolingual.
He also claims that only six parties sit on the Brussels municipal council, but this is factually incorrect: there are nine, since parties that ran on joint lists still operate separately once elected. This blurring of lists and parties is not innocent — I return to it below. Naturally, more parties make coalition-building harder, but they also make for a parliament that better reflects the diversity of the population.
A perfect political storm
Second, Van Parijs himself observes that the seven-party coalition eventually formed had effectively been on the cards from day one. Would it not be fairer, then, to conclude that some of the protagonists lacked political skill, while others deliberately dragged out negotiations to arrive, 613 days later, at precisely that outcome? Those 613 days were twice as long as every previous Brussels coalition negotiation combined since 1989 — normally, the same parties strike a deal within one or two months.
The 2024–2026 episode looks less like proof that the system has failed than like a perfect political storm. Municipal elections were looming, the federal coalition talks were themselves fraught, and several parties were fighting internal power struggles.
Some politicians dragged their feet simply because they could, others because they judged it necessary to change course: after five years of regional government that left a budget deficit reaching 25%, a hasty restart was never realistic. Voters, for their part, had made their frustration clear by sending both the French-speaking and Dutch-speaking outgoing coalitions packing.
Is the Dutch-speaking minority really overrepresented?
Third, we come to the numbers game over how many Dutch speakers actually live in Brussels. As a member of that minority, I find it dispiriting to see it, time and again, whittled down to near-nothing through partial statistics that tell only part of the story. To be fair, a small language like Dutch has a hard time asserting its relevance in the most international city in Europe, dominated as it is by two of the world's most global languages, French and English.
Van Parijs cites figures showing a decline since 2000. What he omits is that Dutch usage has actually been rising over the past five to ten years — visible in the number of Dutch speakers registering at the municipal level, the growing influx of Flemish residents into Brussels, and the continued expansion of Dutch-language education at every stage, from kindergarten to adult learners. This is no small feat in a city that draws migrants from across the globe who typically pick up French or English as a second language, but rarely Dutch.
This raises the question of whether Dutch speakers are really as politically overrepresented as Francophone politicians often suggest. Georges-Louis Bouchez, president of the MR, frequently claims that only 5% of Brussels residents speak Dutch — a figure that is both wildly inaccurate and beside the point.
It is inaccurate because, as Van Parijs correctly notes, the true figure is 12%, a remarkable feat given that Brussels' population has grown by 30% over the past three decades, driven largely by international migration. And it is beside the point because what matters is the share of Dutch speakers within the Belgian electorate, since only Belgian nationals can vote at the regional level.
A study by the (Francophone) ULB puts that figure at 15.8%, only slightly below the 19.1% of seats the Dutch-speaking minority holds in the regional parliament. Since that share is still growing — while Brussels' population has stopped expanding — the “overrepresentation” may disappear entirely within five to ten years.
Tactical voting is nearing its limits
Fourth, Van Parijs makes a genuinely valid point: Brussels residents no longer vote along strict linguistic lines, as they did when the Region was created in 1989. Extrapolating from a ULB study, roughly 12,000 Dutch speakers voted for French-speaking parties, and around 34,000 French speakers voted the other way. A degree of “arbitrage” between the two electoral colleges is indeed possible, as Van Parijs notes, since each Dutch-speaking vote currently carries fractionally more weight than a French-speaking one. But that arbitrage is likely to hit its ceiling by 2029.
If the Dutch-speaking vote grows by another 10,000, its seats will no longer be “cheaper” than French-speaking ones. What would remain is the gap in thresholds between the colleges — roughly 4,500 votes on the Dutch-speaking side versus 19,000 on the French-speaking side — which is what tempts new parties to run in the smaller college, as Agora did in 2019 and Team Fouad Ahidar and Viva Palestina did in 2024, despite drawing mostly French-speaking voters. One remedy would simply be to lower the French-speaking threshold to match the Dutch-speaking one.
That might, as Van Parijs fears, fragment the French-speaking seats further — but perhaps that extra competitive pressure is exactly what Brussels' democracy needs, after five years of dismal budgetary performance and 613 days of political theatre. True democrats do not fear new parties; they welcome them as a possible breath of fresh air.
All of this leads Van Parijs to conclude that bilingual lists are the logical, inevitable and only way to reform the Brussels electoral system. They would sharply cut the number of lists, ease coalition-building, and eliminate arbitrage — all true enough. But this glosses over one crucial fact: it would effectively end the autonomy of Flemish parties in Brussels. Van Parijs does favour preserving the guaranteed representation of 17 Dutch-speaking MPs, but he overlooks how easily a system of bilingual lists could be exploited to hollow out the political weight of Flemish parties and candidates alike.
The 1971 precedent: “faux flamands”
To see why, we need to revisit 1971, the year of the first and only election to the Brussels Agglomeration Council, forerunner of today's regional parliament. Bilingual lists were allowed then too, with fixed quotas for French- and Dutch-speaking candidates. But the Francophile FDF gamed the system by placing so-called “Faux Flamands” — false Flemings — on its lists, candidates who simply declared themselves Dutch-speaking on their identity card while remaining Francophone in practice. A change of ID card was all it took. In doing so, the FDF filled the quota reserved for Flemish candidates and subverted the entire system.
As a result, the Agglomeration Council never functioned properly and stood paralysed for eighteen years. Van Parijs' proposed system risks reopening the door to exactly this kind of manoeuvre. The current arrangement is far from perfect, but it at least protects Flemish parties through a separate electoral college. What is more, the sharp rise in Dutch-speaking votes — from 50,000 in 2014 to 80,000 in 2019 — has pushed the 5% threshold up considerably, from 2,500 to 4,000 votes, making it much harder for new “false Flemish” parties to clear the bar. Under the 2024 threshold, Agora, which won a seat in 2019, would not have succeeded in 2024.
The rise of Team Fouad Ahidar, which draws mainly on French-speaking voters, does pose a fresh threat: it holds 3 of the 17 Dutch-speaking seats. Tellingly, Fouad Ahidar spent twenty years as a Vooruit (Flemish socialist) politician before striking out alone, and remains legally obliged to run in the Dutch-speaking college. Further growth could eventually threaten the system, but it would need to more than triple its vote count to secure an absolute majority.
Back in 2004, Vlaams Belang came close to a majority in the Dutch-speaking college, drawing heavily on French-speaking protest votes over safety concerns, yet still fell short with 6 of 17 seats. Given the subsequent growth in Dutch-speaking votes, it would now take at least 40,000 votes to win an absolute majority there — nearly double Vlaams Belang's historic peak in Brussels. Can Team Fouad Ahidar get there? Time will tell, but in any case, the Brussels electoral system cannot be changed before 2029. It is also worth noting that the party has since lost a quarter of its municipal council seats to defectors.
Bilingual lists would currently wipe out the Flemish parties
Given the current number of Dutch speakers in Brussels, bilingual lists would today all but eliminate the influence of Flemish parties. Going it alone would never be a credible BATNA (best alternative to a negotiated agreement), since their Francophone counterparts could safely assume that any Flemish party running solo would fail to clear the 5% threshold — as Van Parijs himself concedes. What leverage would that leave them?
It is true that at the federal and municipal levels, sister parties have almost always joined forces, but only because the Flemish parties know they cannot win seats on their own. Tellingly, those joint federal lists did not even carry the Flemish party names — they ran simply as PS, MR, Ecolo and Les Engagés. That is a fair preview of what bilingual lists would mean at the regional level too: the Flemish parties would become, to borrow the French phrase, a quantité négligeable — utterly irrelevant. As for N-VA and Vlaams Belang, they would stand no chance of winning any seats at all, meaning the country's two largest parties would go entirely unrepresented in the parliament of the Brussels-Capital Region.
It is worth remembering that Brussels' current electoral system was designed at the federal level, not the regional one. Belgium's balance of powers is delicate and deeply interconnected. A wholesale overhaul of the Brussels system could therefore open a Pandora's box. In my view, there is certainly room for modest adjustments, but a complete overhaul would put Belgium's own stability at risk — reopening questions such as ministerial parity at the federal level, the special protections for the Francophone minority nationally, and the status of the communes à facilités (Belgium's linguistically mixed “facility” municipalities).
Minor modifications are preferable to a complete overhaul
I see three modest changes that could ease the current tensions without a full overhaul. First, as noted above, the electoral threshold should be equalised across both colleges. This would remove the incentive for parties drawing mainly on French-speaking votes to run in the Dutch-speaking college purely to subvert the system. Yes, this might further increase the number of parties in the Brussels parliament, but it would also put pressure on the incumbent French-speaking parties, whose track record over the past five to ten years has hardly been stellar.
Second, a language requirement should apply to anyone elected in either college. It is simply unacceptable that one of Team Fouad Ahidar's MPs, elected to represent the Dutch-speaking community, cannot speak a word of Dutch. (That MP, incidentally, is currently facing legal trouble and may soon lose his parliamentary immunity.) There should be no restriction on who may stand as a candidate — that would be unconstitutional — but once elected, an MP can only properly represent a linguistic group whose language they actually speak. Constitutional experts I have consulted believe such a targeted requirement would withstand scrutiny.
Third, it would make sense to introduce a degree of “floating” between the two colleges' seat allocations. The current 72–17 split is entirely fixed, but it could instead be allowed to shift within a range — say, between 78–11 and 66–23. This would discourage French-speaking voters from casting tactical votes in the Dutch-speaking college, since doing so would boost Dutch-speaking representation, which is unlikely to be their goal. It would also keep the number of votes required per seat roughly comparable between the two colleges, reducing the sense of injustice created when Dutch-speaking seats come cheaper than French-speaking ones, as they do today — though, as noted, that gap may vanish on its own by 2029.
Should EU citizens vote in Brussels' regional elections?
Finally, there is the question of EU citizens' exclusion from regional elections. Some — mostly Dutch-speaking — politicians have long argued that enfranchising them would add roughly 250,000 voters and shrink the democratic deficit. Van Parijs, too, clearly favours extending the vote to EU citizens. My own view is more nuanced.
First, Belgium still has compulsory voting at the regional level. Extending the franchise as things stand would therefore also compel 250,000 EU citizens to vote, on pain of fines for those who fail to do so — hardly a welcome prospect for most of them. The first change needed, then, is to replace compulsory voting with a right to vote, as is standard practice almost everywhere else in the EU.
Second, it would make little sense to enfranchise EU citizens who have lived in Brussels for only a few months, or even years, without any real intention of staying. I would set a minimum residency requirement of ten years. That would ensure new voters have a genuine long-term stake in the region, a solid grasp of its political landscape, and the means to cast an informed vote. Under such a rule, the pool of new voters would likely be closer to 25,000 — a far less disruptive number, particularly set against the roughly 500,000 votes currently cast at the regional level. As a benchmark, only 16% of eligible EU citizens currently bother to vote in Brussels' municipal elections.
Third, if Brussels opens regional voting to long-term EU residents, should that right not be extended across the EU for all regional elections? Why should a long-term EU resident of Brussels gain a democratic right that a long-term Belgian resident of Berlin or Madrid would be denied in elections for the Bundesland of Berlin or the Comunidad Autónoma of Madrid? I would go further and argue that such reciprocity is required to avoid discrimination — meaning this question should be settled at EU level, just as it was for local elections. Politicians who claim it can be fixed unilaterally and overnight are likely playing to their more populist instincts, hoping to court Belgian voters with EU roots or dual nationality.
Taking the long view: how bilingual lists could become reality
To conclude, Van Parijs' proposal has real merits and could well strengthen Brussels' regional democracy — but its time has not yet come. It would virtually erase the political clout of Flemish parties in Brussels, which is not only unacceptable to those parties but arguably also against the best interests of both the Brussels Region and Belgium as a whole. It risks accelerating Flemish disengagement from Brussels, and even paving the way toward outright Flemish independence.
Paradoxically, Van Parijs' proposal only becomes viable, in my view, once the share of Dutch speakers within Brussels' Belgian electorate rises to around 30% — roughly double today's figure. At a growth rate of 0.5% a year, that could take some thirty years. At that point, Flemish parties would have real leverage in negotiating joint lists with their Francophone counterparts: a credible go-it-alone option, and a potential loss of votes for their partners large enough to secure a fair balance of power within joint lists.
It would also allow parties like N-VA and Vlaams Belang to win seats of their own, though the threshold might still need lowering to around 3% to bring them on board. For now, reforming Brussels without the consent of both sides — in practice, without N-VA's consent — is simply not politically realistic.
Forcing the issue could well push N-VA and Vlaams Belang to join forces at the Flemish level to seize power together, which would in turn deadlock the federal parliament and open the door to an existential political crisis for Belgium itself. Be careful what you wish for.
