Brussels has a daytime population substantially larger than its resident population. A very large number of people commute in each morning from Flemish Brabant, Walloon Brabant and further, use the roads, the transport network, the policing and the emergency services of the capital region, and go home to a commune that collects their personal income tax.
The mismatch is structural rather than accidental. Belgian personal income tax is levied where a person lives, and a share of it is redistributed to that person's commune. A city whose workforce lives outside its boundaries therefore carries the cost of the infrastructure that workforce uses without collecting the revenue that workforce generates, and the effect compounds when higher earners are the ones most likely to move out.
Housing is what drives the movement. Brussels property is expensive relative to the surrounding provinces, family sized housing with a garden is scarce inside the nineteen communes, and the regional differences in registration duties and housing incentives make the calculation more complicated than a simple price comparison. Families move out at a predictable point in their lives, and often keep working in the city.
Successive proposals have tried to address the revenue side, from a commuter levy to a broader reform of how the regions are financed, and none has passed. The obstacle is that any fix transfers money from Flanders and Wallonia to Brussels, which makes it a question about the federal settlement rather than about municipal accounting, and questions about the federal settlement move slowly.
What has changed instead is behaviour. Remote working has cut the number of days commuters spend in the city, which reduces pressure on the transport network and also reduces the spending those commuters do at lunchtime and after work. Brussels has traded one problem for a different one, and the underlying arithmetic is untouched.