Thoughts on Quadratic Voting

Posted on December 8, 2018
> Epistemic status: I read the Liberal Radicalism Paper and the book Radical Markets, 
and formed some ... opinions.

If one is involved in the Ethereum space, the name Glen Weyl has recently popped up. This is because firstly, his book, co-written with Eric Posner Radical Markets is popular among the Ethereum set. Secondly, Vitalk and Weyl recently co-wrote a paper called Liberal Radicalism that has proven to be a popular hit.

While there are many interesting ideas espoused in the book and the paper from the idea of property and monopoly to data as labor, the idea of Quadratic Voting struck out to me in particular.

For a brief introduction, Quadratic Voting is a decision-making method where people allocate their preference for various decisions from a pool of voter credits. However, while one can allocate additional votes, the amount of credits used per vote is exponentially, so voting strongly on an issue can quickly become more expensive.

For reference, here is a table to see how this would play out:

Actual Votes Vote Credits Spent
1 1
2 4
3 9
4 16
5 25

Given a limited pool of vote credit, if you feel strongly about a single issue you can plow your votes into that issue. However, other issues you might like might not get passed because you ignored them, and others voted strongly for them.

The examples given to me at least at the RadicalxChange meetup in Berlin would be that Quadratic Voting would have prevented Trump from being elected.

While I believe Quadratic Voting at minimum is an interesting idea. I do have some trouble with the way people are going about it.

Firstly, while this reflects the politics of the sub-culture, it feels like some people are taking Quadratic Voting as a salvation against views outsider their circle. Quadratic Voting might actually accentuate more partisan Voting, as the dedicated base of partisans might plow their votes into getting their candidate into office. An example of this is if Quadratic Voting existed in Brazil, Bolsanaro likely would have gotten an even larger share of the vote. Considering the discontent Brazilian voters had which propelled Bolsanaro into office, this is a fair prediction.

Secondly, there was an example given why Quadratic Voting is good that I find troubling. An example was that minority groups could use more of their votes to defend against laws/proposals that target them. This has worrisome implications because it implies minorities will be heard from less because they are busy defending themselves against targeted proposals. While someone involved in the idea of Quadratic Voting thought this might be mitigated because everyone is a minority in some way, to butcher Orwell, some people are more of a minority then others.

Also, if you are of the bent that direct democracy, such as having a lot of propositions or referendums is not a good idea, constantly Voting on issues directly instead of delegating your votes to someone this system might not be as efficient.

Quadratic Voting and delegation could be combined.
It would be funny to see congresspeople have vote pools that look like some weird spider chart when represented. 

Also, as it is stated to be based on market principles, this troubles me a little bit. While the free market often times can be a useful tool, I personally believe sometimes the free market has to the submissive to the interests of the common good, and the market should serve the public and not the other way around. The idea of the marketization of Voting then is worrying, considering due to lobbyists and other groups politics is pretty marketized already.

For reference, this article by Epsilon Theory more a less tracks my opinion on people sticking the market in places where it might not be the greatest idea.

My trouble with Quadratic Voting is, while I disagree with the idea somewhat, I do not want to dismiss it out of hand because frankly it is a refreshing idea on how to vote, and we need fresh ideas especially now. Quadratic Voting at minimum, deserves to be taken into equal consideration like Georgism and LVT, strange, but definitely has some aspects that could prove useful in the long run in fixing deep seated issues.

Off to the idea mines, I guess.