Optional Asset Voting

The idea would be this: keep Plurality, but let voters “circle” a candidate (or check a box at the top of the ballot) to allow this candidate to trade their vote. This would allow voters to choose weaker candidates without fear, and if the candidates are honest, eliminates vote-splitting. There would be less risk of corruption, because voters make the decision to make their vote tradeable or not, not the candidates.

Example: Five candidates in a race. 20% of the voters feel coerced to pick one of the two frontrunners under Plurality. Under OAV, they simply mark the box for vote-trading, and their votes ultimately still count. Over time, perhaps the minor candidates start to get elected.

This is a less complete solution than the ones we’re looking at, but I think it could have a huge benefit over Plurality at little cost.

It might even be more possible to get this passed in certain jurisdictions than other systems (because of the minimal change required), but that’s just a wild guess for now.

It also resolves much of the trust issues that voting reformers and the public have around Asset, as only a voter who has strong trust in their candidate (or who has been observing the political process keenly) is likely to permit vote-trading for their candidate.

There could be two implementations: a checkbox at the very top of the ballot, permitting or not permitting every candidate to trade your vote, or a checkbox/circle thingy for each candidate you vote on, or something in between.

Edit: Maybe you could vote for one candidate, while letting another transfer your vote? Seems like it’d have security and legitimacy issues, but might have useful applications too.


At least at first blush, I could see that as better than the current system.

It does have the risk of voters forgetting to check boxes, or miscalculating whether to check the box, but this is definitely more viable than full Asset. Probably a bit “closer” conceptually to Plurality as well.

EDIT: This is totally invalid. You create 2N candidate entries: AliceTransferable, AliceStatic, BobTransferable, BobStatic, etc.
The problem is that is makes it harder to use dumb totalizing machines. Such machines would simply report something like

NoTransfers: 473
YesTransfers: 622

A: 200
B: 58
C: 610
D: 227

(Approval and Score single winner work on DTMs.)

Thanks for anticipating and fixing it. I do agree you lose the minimal change to the ballot, since now you’re required to put checkboxes and instructions in there, but it’s worth it for most folks who get worried about full Asset; we have to start somewhere to get some better voting systems.

As it would be irrational to opt to have your vote be wasted in the absence of a majority winner, the only benefit I can see to allowing the voter to opt out of asset voting altogether would be to reduce the power of irrational voters.

I do, however, really like the idea of allowing the voter to opt to let a candidate other than his first choice transfer his vote.

That would be especially useful in single-winner elections, where the winner won’t have to compromise: the priorities of the candidate you would prefer to be the winner (i.e. dictator) may not align as well with yours as those of another candidate; he might be the ideal winner (i.e. dictator) but not the ideal transferrer (i.e. compromiser).

And while the probability of that being the case diminishes rapidly as the number of seats (and thus the necessity for winners to compromise) increases, it may still be useful to generalize optional asset voting to the multi-winner case, provided the default owner of an exhausted vote is its first choice. Where there are multiple first choices, as in basic approval, the appropriate default owner would depend on the ballot layout: if one lists all candidates one approves of, the first listed candidate should be the default owner; if there are pre-listed candidates one marks approval of, an exhausted vote should be divided among all approved candidates unless the voter identifies his preferred owner.