Mechanism Design: Governance Protocol & Voting

Problems in Voting Mechanisms Today, What Solutions Tomorrow

Government and politics aside, voting itself is inherently flawed when we add in the irrationality of human beings and the short-term focus as individuals. This is why we need robust governance models to achieve efficiency in the system. In mechanism design, voting is a very crucial slice of pie that people seem to take lightly.

Sure, with everyone given a voice to vote on an issue, we can reach the democracy that the Greeks dreamt about years ago. But with a voice per vote, it comes with a lot of other complications.

Fantastic read if you love voting: The Ethics and Rationality of Voting.

In the blockchain space, some issues are:

1. Lack of proper incentives by the system
2. Punishments or de-incentives for bribery (on-chain or off-chain)
3. Impact of repeated voting on the outcome of votes
4. Rationality of voters
5. Externalities like social contracts that can impact the short-term outcomes.

This is a post on the issues/challenges regarding voting, what is the current development & what further research is NEEDED in the space.

Note: voting is a general term. Depending on the topics like voting for president to voting for “what’s for lunch tomorrow”, the mechanisms are different, and so are the incentives.

Issues/Challenges regarding Voting

Issues with Voting (Centralised or Decentralised)

• 1 vote on Yes/No is not an accurate measure of people’s preferences. Polarised views and extreme preferences are not taken into account and this could affect results.
• Quality of decentralised/crowd-sourced votes because of issues like misalignment of incentives, voters are not experts on the issue and voters are more short-term focused compared to the change-makers (i.e. relationships between citizens and government, customers and business owners).
• Centralised entities now to give up their power. There is no incentives for them to adopt the new decentralised system. (Let’s face it, humans love power.)
• Large scale voting where everyone needs to vote is a repeated game. With repeated games, it is important to look beyond the single transaction and consider the long-term incentives for this ongoing social interaction.
• Incentives or mechanisms that teases out the socially optimal choices instead of individual gains. (Think Vickrey-Clarke-Groves auction but for voting.)
• Our current system does not align to transparency and accountability (I mean , look at Trump), and changing the fundamental values of the society/system is a big step.
• When we look into decentralisation, we are also distributing responsibility to every node in the network. This means people have to be more accountable for their actions and baggage that comes with the decision. There also needs to be a focus on education and learning how this new system works. Now, THAT’s a huge challenge.

In Business Language

• Plenty of issues with the current system of voting. The current system does not truly reflect our preferences and best choice for society as a whole. Example: bribery (I pay you £X and you vote for Option 1).
• Also, there will be a shift in the mindset of people as we move to a decentralised system with power and responsibility more evenly distributed.

Current Development in Voting Space

Current Development in Voting Mechanisms

• Quadratic Voting: Alternative to one-person-one- vote rule. Voters can vote as much as they want, and pay the square of the number of votes. (E.g. cast 10 votes, costs 100-currency.) This pricing concept has a solution that benefits everyone in the network. As more people vote under this system, the better the results are.
• Liberation Through Radical Decentralization: Ethereum is working with Quadratic Voting to address some issues. (E.g. gauge community sentiment on changing protocols.) But there are issues like anonymity and pseudonymity in the crypto-space. Issue of reestablishing monopolies may arise.
• Quadratic Vote Buying, Square Root Voting, and Corporate Governance: example of how Quadratic Buying is applied to corporate governance. In this case, they created a simpler model, Square Root Voting (SRV) for corporate governance. This method should also give shareholders a stronger incentive to monitor corporations and give corporations a stronger incentive to hold votes than they have under the existing system. This system also protects minority shareholders directly by making the voting rule resilient against abuse and manipulation.
• How Mechanism Design Can Radicalize Democracy (Quadratic Vote Buying): 1. Individuals receiving votes equal to the square root or radical of the voice credits they spend. 2. True modern democratic institutions: ensure the state serves the general happiness of the people maximally. 3. Radically expands the rights of citizens to fully and freely express their political views, liberating them from the straight jacket of 1p1v influence rationing.
• Quadratic Voting as Efficient Corporate Governance (Minority shareholders): Current issue in corporate governance in shareholder voting is inefficiency in the system. QV is a mechanism design that allows for efficient corporate governance because it allows people to share their true feelings (private valuation) to the issue.
• Research on voting by E. Glen Weyl: More work done by E. Glen Weyl including pricing theory, economics of labour market and more.
• DFinity, Tezos, Futarchy, Quadratic Voting — Solutions to the Blockchain Governance Problem?: How different blockchain platforms use on-chain governance (mechanism design)
• Interview with Glen Weyl on radical markets: Glen Weyl talks about how his work with Vitalik and experiments on QV in blockchain
• Decentralized Voting: A Self-tallying Voting System Using a Smart Contract on the Ethereum Blockchain: A secure decentralized online voting system using cryptography and a smart contract, which allows the voters to cast their ballots by assigning arbitrary numbers of points to different candidates. They also share a security and performance analysis, showing the feasibility of the proposed protocol for real-world voting applications at large scale.
• PeerVote: A Decentralized Voting Mechanism for P2P Collaboration Systems: Fully decentralized voting mechanism PeerVote, which enables users to vote on modifications in articles in a P2P collaboration system. Simulations and experiments shows that the mechanism still works even with bad actors around.
• Continuity and Incentive Compatibility in Cardinal Voting Mechanisms: Many important models in mechanism design preclude the use of monetary transfers. Paper proves that every cardinal incentive compatible voting mechanism satisfying a continuity condition, must be ordinal. (Will digest this paper again and simplify it.)
• Robustly Coalition-Proof Incentive Mechanisms for Public Good Provision are Voting Mechanisms and Vice Versa: With no participation constraints, the outcome will benefit everyone in the network. However, sometimes these results cannot be achieved because it requires information transparency, and sometimes this hurt the people who provide it. The requirement to eliminate this possibility is called “coalition-proofness”.
• Uncertainty, polarization, and proposal incentives under quadratic voting (QV): QV can provide an incentive for individuals to propose policies with greater uncertainty. The incentives of a potential proposer can come into direct conflict with maximizing social welfare.
• Approval Voting and Incentives in Crowdsourcing: 3 problems with crowdsourced voting: (1) the workers are not experts; (2) different incentives by voters and organisation; and (3) voters cannot convery their knowledge accurately, by forcing them to choose 1 option in the set. Solution: Introduce approval voting. It uses partial knowledge of people + incentive mechanism.

In Business Language

• Plenty of research looking at ways to change our voting mechanism to ensure that everyone has a better voice and reduce social inefficiencies.
• Adding this to the governance protocol (be it a blockchain platform or a DApp), we can create better systems with less inefficiencies.
• Problem: this is still new, lots of testing to be done and parameters to define. Lots of challenges too, but worth looking into.

Possible Further Research Needed

Further Research Needed in Decentralised Systems

• Quadratic voting in a decentralised system and the cost of vote buying (bribery)
• Quadratic voting in an asymmetric Bayes Nash Equilibrium and/or other dominant strategies
• Impact of continuous repeated social contract (or continuous public goods) between voters and implementers
• How efficient is transparent decentralised blockchain systems on accountability and fraud compared to systems today
• Impact of bad actors tipping the decentralised voting system given vote buying/alternative ways of bribery (E.g. how many % of bad actors/bribers can tip/destroy the system)
• Impact on short-term vs long-term incentives of bribery in a decentralised system (i.e. short-term = financial incentives now. Long-term = building social infrastructures for future generation like new schools, hospitals, etc.)
• Impact of short-term view of voters and long-term goals/view of government. How much does decentralised voting (1 person 1 vote) create socially optimal outcomes? (Case in point: government spending a lot in R&D of technology that affects future generation, and issue is voted on by existing generation that do not share the sweet success.)
• Mechanism design for accountability and socially optimal outcome. A VCG version of voting rights, perhaps? (An alternative to QV)
• [Super Future] New markets, issues and solutions given the research of everything above

In Business Language

• We need to design better mechanisms to incentivise people and create a fair system*
• We need to understand the externalities that could affect the voting mechanism design, like hackers or non-financial benefits

Conclusion

Voting in mechanism design is still in its infancy stage when it comes to blockchain. As we open Pandora's box, more problems will start to appear, as is with new technology. The technology will continue to develop, and we, researchers/economists/future leaders, should definitely continue in the direction of new research to solve new problems.

Happy to hear of further developments in the space and/or ideas that will define our future!

Fair system* is a fairly new term just like “human rights”. What is a fair system and how do you define it? That’s another topic for another day on the philosophical definition.