A March 26, 2026 working paper published by the ECB examined governance in Aave, MakerDAO, Ampleforth and Uniswap. The authors found concentrated token holdings and difficulty identifying some influential participants.
The paper expressly represents its authors’ views, rather than an official ECB policy determination.
Related reporting: Parsec closes after five-year run as DeFi and NFT slowdown reshapes crypto analytics market
Holdings and delegated votes have different denominators
The top 100 addresses held more than 80% of governance tokens in each protocol studied. Separately, the paper’s voting table showed Ampleforth’s top 20 voters controlling about 96% of delegated votes, equivalent to 4.8% of circulating supply.
For MakerDAO, the top 10 voters represented about 66% of delegated votes; for Uniswap, the top 18 represented about 52%. These historical sample measurements are not live governance readings.
Identifying accountable actors
An exchange address may represent multiple customers, while several addresses can belong to one entity. Address concentration therefore requires interpretation before assigning ownership or control.
The findings inform debate about identifying relevant actors in DeFi. They do not establish a binding MiCA classification for any of the four protocols.
Our DAO governance guide explains token voting and delegation.
Editor’s note: The article now links to the paper, identifies the findings as historical sample data, distinguishes delegated votes from total supply, and avoids presenting research as an official regulatory ruling.