Candidate-based apparentment (list combination) as an alternative to Asset Voting

Most of this post is meant to offer a solution to this specific criticism of Party List that points towards Asset as a superior simple PR method.

Apparentment is when parties can fuse the votes over the quota they got for their lists into a “super-list” in an attempt to get more representation i.e. a party gets 1.4 of a quota and can’t use that 0.4 of a quota to elect anyone, but if they have made an alliance with another party that got 1.6 of a quota, they can combine their surpluses to elect a representative that they agreed on before the election.
Why not do it with candidates?

Candidates could run independently, but if they don’t get a quota of votes, they could come up with several layers of apparentments, culminating with the party list and any electoral alliances their parties have made. It would probably look something like Candidate>Couple of candidates that agree on an issue>Subfaction>Larger subfaction>Party/Faction>Party’s Electoral Alliance with other Parties>Etc. Unlike Asset, all of the negotiations are handled pre-election, and it might even be possible to allow a voter to decide how many layers of apparentment they want their vote to travel down before exhaustion (or simply exhaust the vote if their candidate can’t win a quota of their own before any apparentments have been activated.) Allowing a candidate to choose what % or how many of their votes should go to particular apparentments (or even putting conditions on how the votes flow, so long as it doesn’t freeze the algorithm) might be a further improvement.

This could be criticized as being byzantine, but candidate negotiations don’t seem very transparent either. It is, at the very least, precinct-summable and based on choose-one voting.

I’d also wonder if something like this can be used to give a majority-winning coalition a few backup seats in case some of their representatives vacate their posts i.e. parties form a mega-alliance before the election, and if the parties in that alliance collectively get a majority, then they can use the backup list in case a representatives from one of those parties has to vacate. It’s possible to require the consent of the legislature to do this, as well as perhaps allowing several electoral coalitions to form backup lists, with the legislature getting to decide which coalition and which list to honor in case one of the parties in that coalition has a vacancy.

So would all winners be selected algorithmically? Who chooses winners within a layer of apparentment? What happens if more than one quota of surplus is scattered at the highest level of apparentment so no single coalition has a full quota? Is it possible for my party to form a coalition with just one faction from another party?

Just to add, it might be possible for a candidate to have several lists or choices of apparent any and let the voter write in which particular one they want to use, for even more choice.

I think this idea I had might be a fascinating way to algorithmically select the winners. Ignoring it though, even STV probably works, no?

I’m probably not thinking at the level you are on this one. Does my STV answer suffice? Is it possible to just pick whoever would have the most votes after adding in that layer of apparentment, provided they have a quota?

That would probably be exhausted, or maybe then and only then Asset could be applied. If Asset is applied, I’d propose allowing the majority coalition (if there was one) to get just enough top-up seats from a list to ensure that there is a coherent majority that doesn’t depend on representatives elected via Asset to govern.

Edit: Actually, I’d be curious whether a candidate could bring their vote back after it’s unable to form something at the party list electoral alliance level. Maybe in their layers of apparentment, they can specify after the vote goes to the party list electoral alliance, it can go back to some other smaller group of candidates, or maybe the method can just do reverse-cycling where it starts to go back down from the highest layers of apparentment to see if it can elect more people.

I think so? Any group or individual could form a coalition with any other group or individual.

What I mean is: suppose I am the leader of the grue party, and I’m negotiating a coalition with the leader of the bleen party. The bleen party has two factions: the bl faction and the een faction. I am willing to help the bl faction, but I don’t want to contribute to the seat count of the een faction. Can I do this?

I think so, but it’d have to be explicitly stated. Actually, why would you even go to the leader of the bleen party when you can just go to the leader of the bl faction?

Also, this idea should offer a natural way to introduce thresholds; if some candidate is required to get 5% of the vote to win, then just keep following the layers of apparentments until that threshold is met.

It’s possible for the candidates to submit rated ballots and then run a cardinal PR method on that with this idea in various ways. Maybe even mix ranked and rated PR methods in various steps of a larger algorithm.
To go a step further, candidates could submit multiple lists, and then the voter could write down which of their candidate’s lists they want or prefer.