Fairness in Token Delegation: Mitigating Voting Power Concentration in DAOs
This paper presents an empirical study of 14 DAOs revealing that current delegation mechanisms exacerbate voting power concentration and misalignment with token holders' interests, arguing for interest-based delegation systems to improve governance fairness and representativeness.
Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Imagine a giant, digital town hall where everyone owns a ticket (a "token") that gives them a say in how the town is run. This is a DAO (Decentralized Autonomous Organization). In theory, every ticket holder should get to vote on important decisions, like building a new park or changing the rules of the road.
But in reality, most people are too busy, too confused, or just too tired to vote every time a new issue comes up. So, they do what many of us do in real life: they hand their ticket over to a neighbor they trust to vote for them. This is called delegation.
The paper you shared, "Fairness in Token Delegation," investigates what happens when people hand over their voting power. Here is the story of their findings, told simply:
The Problem: The "Popular Kid" Effect
The researchers found that the current system is broken in a specific way. When people look for someone to delegate their vote to, they usually see a list sorted by popularity. It's like walking into a school cafeteria where the seating chart is arranged so the most popular kids sit at the biggest tables, and everyone else is pushed to the back.
Because the list is sorted by "who has the most votes right now," the same few people keep getting more votes. It's a "rich-get-richer" cycle.
- The Result: A tiny group of "Super Delegates" ends up holding almost all the power, even if they don't actually care about the same things as the regular people who gave them their votes.
- The Analogy: Imagine you give your vote to a celebrity because they are famous, but you care about fixing potholes, and the celebrity only cares about building a new stadium. You've handed your power to someone who doesn't represent your interests.
The Study: Connecting the Dots
To prove this, the researchers acted like digital detectives. They looked at 14 different DAOs (like Aave, Uniswap, and Compound). They did two main things:
- Linked the Real World to the Digital World: They matched people's usernames on discussion forums (where people chat about what they care about) with their digital wallet addresses (where they actually cast votes).
- Read the Room with AI: They used advanced AI (Large Language Models) to read thousands of forum posts. The AI figured out what topics each person cared about—like "treasury spending," "security," or "new technology."
The Big Discovery: A Mismatch
Once they had this data, they compared the interests of the regular voters with the actual voting history of the delegates they chose.
- The Finding: There was a huge mismatch. People were often delegating their votes to delegates who had totally different priorities.
- The Metaphor: It's like a town where the residents are mostly worried about "keeping the lights on" (security and budget), but they keep voting for a mayor who is obsessed with "building a space station" (high-risk experiments). The mayor gets all the votes because they are the most visible name on the list, not because they agree with the town's needs.
Why This Happens
The paper points out that the tools we use to find delegates (like a website called Tally) make this worse. These tools show a list of delegates sorted by how many votes they already have.
- The Bias: Just like how products at the top of an Amazon search result sell more simply because they are at the top, delegates at the top of these lists get more votes just because they are visible. This creates a feedback loop where the powerful get more powerful, regardless of whether they are actually good representatives.
The Proposed Solution: "Interest Alignment"
The authors suggest a new way to organize these lists. Instead of sorting delegates by how many votes they have, we should sort them by how well they match your interests.
- The New Analogy: Imagine a dating app for voting. Instead of showing you the "most popular" people, it asks: "Do you care about security?" and "Do you care about low fees?" Then, it introduces you to delegates who have voted on those specific topics in the past.
- The Goal: This would break the cycle of the "Super Delegates" and give a voice to the quiet majority who care about different things. It would make the town hall actually represent the people, rather than just the loudest voices.
Summary
In short, the paper argues that while delegation was supposed to help regular people participate, the current system accidentally creates a small elite class that controls everything. By using technology to match voters with delegates who actually share their values (not just their popularity), we can make these digital communities fairer and more representative.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.