Collusion's another thing all together, and not the thing you're replying to. Complicated, as always, but in short the scenario you describe might very well be illegal.
Here is some back of the envelope math that I did:
Assume you have two groups who are each 100% focused on a cause which the other group has no opinion of.
n = people in each group
b = individual vote budget
Each Group Votes Independently: n * sqrt(b) votes
If a group colludes s.t. group A gives (1/2)b votes to group B's issue and vice versa you get: 2 * n * sqrt((1/2)b)) votes which simplifies into Sqrt(2) * n * sqrt(b).
Yet coordinating between these two groups becomes more difficult and costly, growing somewhere between n (perfect coordination) and n^2 (total decentralization). Thus, regardless of how you coordinate it will always be more expensive than the influence you gain which grows at sqrt().
Even in this simple assumption collusion becomes a negative value decision.
Collusion is a secret cooperation or deceitful agreement in order to deceive others. The plan you suggest is not deceiving anybody. It represents an political agreement and cooperation.
Alice and Bob are making a good deal. Well working democracy should involve maximum amount of deliberation and deal making. It's a good thing. If you want more of that kind of deal making votes should be visible though (quadratic voting between representatives in parliament might work the way you suggest).
You still would need just 1 participant, it's the computation part that would be open to any number participants in order to reduce the possibility of collusion.
Some people don't like lying about what they are doing, even if they can't get caught. Small scale cooperation between honest people would probably work for that reason. On a large scale, it probably wouldn't be very efficient, but insofar that both sides would defect or cooperate at roughly the same rate, you would still expect it to be somewhat beneficial if the cooperation rate is nonzero (but maybe not enough to be worth the trouble).
reply