The traffic light coalition dares something. She has now submitted a proposal as to how she envisions the reform of the electoral law so that the Bundestag does not keep growing. The number of parliamentary seats is to remain fixed without reducing the number of constituencies. The traffic light has thus made a breach. Big kudos for that.

The Bundestag is known to be oversized. 736 MEPs are simply too many when the law states that the “normal size” is 598 seats. The cause of this miraculous multiplication – first overhangs, then compensatory mandates – has been known for many years. In this way, the back benches continue to be filled, without any real technical or democratic added value. There have already been a few attempts at reform, most recently that of the grand coalition. As is well known, it failed miserably. One has to state that the current electoral system cannot be repaired.

And now the traffic light wants to regulate it again. First of all, she approaches the reform with some caution. The coalition is not presenting a fully formulated draft law; instead, three MPs are drafting a paper outlining what a new electoral law could look like. There’s going to be quite a debate about that. But the Social Democrat Sebastian Hartmann, the FDP man Konstantin Kuhle and the Green Till Steffen have achieved at least one thing with their proposal – they have given the reform discussion a new, better direction.

Because with this model two determinations are made, behind which no one can go back. On the one hand, the Bundestag would always and reliably consist of 598 MPs in the future. It would have a fixed size, which should be the norm in a parliament. No more swaying up to 800 seats or more as previously possible. A fixed number as a target, and unconditional at that – that’s good and right, because part of the messing around with the reform so far has been due to not making that simple prefix.

Second, the proposal avoids the deadly controversy over the abolition of constituencies. It has weighed on the electoral reform debate so far like no other point. Now it would remain at 299 constituencies. The guerrilla war among MPs over the direct mandate applications that would have existed if the number had been significantly reduced would be off the table. That’s a good thing too.

But now comes the big but. This model is not above all doubts either. There are already some big question marks behind it. Even if the three traffic light men do not like to hear it because they are trying to introduce new wording into the debate: what they are proposing is ultimately the well-known capping model, which avoids overhangs (and thus growth) by using the appropriate number Direct mandates are not assigned. Whether that is constitutionally correct will be part of the debate, and it is now primarily about legal issues.

However, the proposal has consequences that go beyond the purely legal. The traditional model of personalized proportional representation, used for the Bundestag since 1949, combines majority voting in constituencies (first vote) and proportional representation system (second vote for the lists). The traffic light suggestion takes care of that, no matter how elaborate the wording gets around it. Since forecasts are now being made down to the lowest level before elections, those constituencies in which caps are being cut should be easy to identify. There, the election campaign takes place under different conditions. Applicants who are “at risk of capping” may be disadvantaged. Equal opportunities across all constituencies is therefore questionable.

But the traffic light proposal is not the only way to permanently guarantee 598 mandates without abolishing constituencies. There are a number of other solutions. As always with the right to vote, the same applies to everyone: There is no perfect system. Everyone has their pimples and warts. And no electoral system in the world is without quirks. The electoral law commission of the Bundestag now has the task of using the traffic light model to check whether there is something better.