Table of Contents
Fetching ...

Blockchain Governance: An Empirical Analysis of User Engagement on DAOs

Brett Falk, Tasneem Pathan, Andrew Rigas, Gerry Tsoukalas

TL;DR

The paper investigates on-chain governance in four major DeFi DAOs by collecting Ethereum vote histories and mapping voters to real-world actors. It introduces the minimal quorum concept, defined as the smallest subset $\{v_1,\dots,v_t\}$ with $\sum_{i=1}^t w_i \ge \tfrac{1}{2}\sum_{i=1}^n w_i$, to quantify power concentration and finds that a small set of top holders can typically decide outcomes. It shows that engagement is low (about 5% of total token supply votes per proposal) but that the costs to sway votes are very high, implying robustness against manipulation; it also maps voters to entities via ENS, Sybil.org, and governance APIs, revealing a dominance of venture funds, university clubs, and founders in voting power, with Lido voters largely anonymous. The results highlight a tension between perceived decentralization and actual power concentration, with implications for governance design, regulatory risk, and future protocol development.

Abstract

In this note, we examine voting on four major blockchain DAOs: Aave, Compound, Lido and Uniswap. Using data directly collected from the Ethereum blockchain, we examine voter activity. We find that in most votes, the "minimal quorum," i.e., the smallest number of active voters who could swing the vote is quite small. To understand who is actually driving these DAOs, we use data from the Ethereum Name Service (ENS), Sybil.org, and Compound, to divide voters into different categories.

Blockchain Governance: An Empirical Analysis of User Engagement on DAOs

TL;DR

The paper investigates on-chain governance in four major DeFi DAOs by collecting Ethereum vote histories and mapping voters to real-world actors. It introduces the minimal quorum concept, defined as the smallest subset with , to quantify power concentration and finds that a small set of top holders can typically decide outcomes. It shows that engagement is low (about 5% of total token supply votes per proposal) but that the costs to sway votes are very high, implying robustness against manipulation; it also maps voters to entities via ENS, Sybil.org, and governance APIs, revealing a dominance of venture funds, university clubs, and founders in voting power, with Lido voters largely anonymous. The results highlight a tension between perceived decentralization and actual power concentration, with implications for governance design, regulatory risk, and future protocol development.

Abstract

In this note, we examine voting on four major blockchain DAOs: Aave, Compound, Lido and Uniswap. Using data directly collected from the Ethereum blockchain, we examine voter activity. We find that in most votes, the "minimal quorum," i.e., the smallest number of active voters who could swing the vote is quite small. To understand who is actually driving these DAOs, we use data from the Ethereum Name Service (ENS), Sybil.org, and Compound, to divide voters into different categories.
Paper Structure (18 sections, 7 figures, 2 tables)

This paper contains 18 sections, 7 figures, 2 tables.

Figures (7)

  • Figure 1: In all the platforms we analyzed, the top 3 or 4 token holders could unilaterally decide almost every vote. The numbers inside the bars represent the average number of voters who voted on a given proposal.
  • Figure 2: The amount of tokens (denominated in USD) an attacker would need to purchase to sway a vote, assuming the existing voters did not change their position. The "robustness" of the protocols does not seem to be increasing over time.
  • Figure 3: Because of the off-chain "temperature checks" that precede the on-chain votes, almost all votes pass by a large margin.
  • Figure 4: Investment firms (venture capitalists like a16z) are frequently members of the minimal quorum
  • Figure 5: Participants in the Aave Minimal Quorums. Although many of Aave's voters are anonymous, blockchain clubs have come to be dominant decision-makers on the Aave platform.
  • ...and 2 more figures

Theorems & Definitions (1)

  • Definition 1: Minimal Quorum