Win Conditions
An election’s “win condition” can also heavily impact the nature of a voting system and the type of leadership it promotes.
There are largely four ways win conditions can be structured:
- First past the post: This is a fancy name to describe a winning condition in which the majority vote is gained. The candidate with the most votes may only have 30% of the votes, making it technically just a “relative majority.”
- Threshold vote: A winner is called when a candidate collects votes surpassing some predetermined threshold. People often refer to these as supermajority votes, but we don’t like the term, as “supermajority” specifically refers to thresholds of over 50% while one could theoretically have a threshold of 30% in a system with seven parties (in this case, a winner would have to both get the most votes and be over 30% of votes).
- Consensus: 100% consensus is required by all parties with voting power before a winner can be declared.
- Proportional implementation: By this criteria, an issue is decided in half measures based on the proportion of the electorate that voted on it. This could translate into anything from the number of seats in a congress (if 60% of the electorate voted Republican, then 60% of the congress is made Republican) to how much funding something gets (if half the town wants a school initiative to get zero funding and half wants it to get $1,000,000, it will get $500,000).
We don’t have a strong favorite among these; however, we don’t recommend using the consensus system when the votes of more than three parties matter. When four or more individuals are forced to make a consensus decision, the decision almost always devolves into a simple dominance hierarchy struggle. This issue might be addressed by physically separating the voting parties and not allowing them to communicate through any means other than writing.
We also warn against proportional implementation because it strongly incentivizes individuals to misrepresent their actual positions. For example, if one party wants $5 in funding and another wants $10, the party that wants $10 is rewarded for pretending they want $20 and the party that wants $5 is rewarded for pretending they want $0. Politicians (or factions within an organization) will promote these “false positions,” causing polarization and the formation of extreme opinions within said group.
Quadratic Voting
At the time of this book’s publication, quadratic voting has become a popular topic in crypto (and among other technocratic voting reform advocates). Quadratic voting allows users to express degrees of preference through a vote rather than a binary response (e.g., to pay for additional votes using money or some form of token, or to vote using a sliding scale) which enables voters to express the level of their sentiment on an issue.
The core gimmick of quadratic voting that distinguishes it from generic tally systems is that it raises or lowers the value of a vote the more a person chooses to allocate to any particular issue. It may, for example, cost nine tokens to get three votes on one issue where those nine tokens could have been used to vote nine times if spread across different issues. In a quadratic voting system, an individual gets more say the more they distribute their vote.
Quadratic voting is useful in any voting system where individuals have the capacity to make a sacrifice in order to increase their vote. This is doubly true if the ability to sacrifice is not evenly distributed across a population (e.g., if people can pay for a vote and not everyone has the same amount of money), as normally when these systems allow an individual to sacrifice financially to gain dominance over outcomes, outcomes end up being virtually decided by the wealthiest individuals. Because every incremental dollar sacrificed on a specific vote choice matters less in a quadratic system, there is some level of protection against this. That protection evaporates if a single rich person can pretend to be many poor people, meaning that in online (especially crypto-related) contexts, quadratic voting is susceptible to Sybil attacks (in which one entity pretends to be many to magnify its vote).
Consensus Decision Making
In consensus decision making systems, near unanimous consent is needed among voters in order for an issue to pass. Such systems are often chosen to incentivize extreme conservatism, which explains why consensus-style decision making is used to make updates to the Bitcoin protocol (which uses a 90% vote among miners to implement such changes).
Consensus decision making is also favored by confederacies that fear fracture and experience severe damage when any particular contingent is offended. A prominent example of this can be seen with the Haudenosaunee (Iroquois) Confederacy Grand Council, which used consensus in decision-making requiring a 75% supermajority to finalize decisions.
Consensus-focused governance structures use an iterative process in which updates and edits are added to a proposal until it no longer has any opposition. This can be achieved either by improving the thing that is being voted on or adding “fat” to the opposite side of the scale (e.g., if a law hurts farmers, it may also include a tax break for farmers).
When people create these systems, they often believe they are trading a slower-to-evolve system for one that will be more deliberative and less likely to offend its constituents. They are wrong.
Consensus-focused systems often drive deliberative bodies to satisfice (a combination of the words “satisfy” and “suffice”) and accept low-risk, easy solutions rather than search for the best solution. Consensus voting systems often rely heavily on social pressure to coerce the minority opinion into agreeing. Typical (total) consensus votes are really decided by 70% or so of the group and the last 20% is achieved through aggressive “social pressure” (bullying).
This process causes the system to “perfect” its methods of social pressure and foster a culture of hostility toward minority opinions, which is typically toxic to a system over the long run. Somewhat ironically, even when a system is not naturally prone to fracture, this governance model makes fracture likely by frustrating fractions of its membership, as can be seen in Bitcoin protocol updates with Bitcoin Cash breaking off during such a deliberation in the process of its few “hard forks” (Bitcoin Cash is an older iteration of the bitcoin protocol run by individuals who never shut it down and moved to a new 90% consensus-driven iteration).