Seems like a recipe for horse-trading. As an abstraction, say we have a group of 400 fairly development-minded people who are mostly indifferent or OK with pro

Seems like a recipe for horse-trading. As an abstraction, say we have a group of 400 fairly development-minded people who are mostly indifferent or OK with projects A, B, C, and D, with the exception that each 25% of this group really likes one of these projects (the classic single-issue voter). Without horse-trading we get them spending their 100 points on their pet issue, getting 10 votes per voter, yielding 1,000 for each. With horse trading, they agree to split their votes with all 4 of the cabal's issues, which means that they can each contribute 25 points to 4 different projects, or 4x5 votes per voter, yielding 2,000 total. So a disciplined party 'bloc' can exert outsized control in this scheme. Of course, the usual problems of maintaining discipline arise, especially with secret ballots where you can't prove that you kept faith with the cabal. Another element of this (and any scheme where people's ability to express preferences is "rationed") would be the issue of what makes it on to the ballot to begin with - there's obviously some real prospects for mischief and vote-splitting. If you have the resources, the idea of creating a lot of moderately trivial proposals to draw away votes from things that you're opposed to might be effective - so if you want to get some huge development through that's opposed by environmentalists, you might attempt to jam the zone with a gazillion minor environmental proposals (all of which look fairly good) to try to suck away voting power from your political opponents.