Governance as a feature

The evidence for the framing is in the code

The base fungible standard has no concept of voting at all. ERC-20 specifies a transfer interface: total supply, balances, transfer, approval and allowance, with Transfer and Approval events.¹ Nothing in it tracks who held what at which block, which is the accounting a vote needs.

Voting arrives as an extension. OpenZeppelin's ERC-20 documentation lists ERC20Votes, support for voting and vote delegation, in the same optional extension list as ERC20Capped for supply caps, ERC20Pausable for halting transfers, and ERC20Burnable.² It sits at the same level as a pause switch. That is the whole argument in one line of documentation: a project that describes itself as launching a governance token has described adding one module.

This matters commercially because features are additive and categories are not. A payment token can carry a vote. A staked position can carry a vote. Restricted, permissioned instruments can carry a vote. Treating governance as a separate species blocks the useful question, which is where to attach it.

What the feature is actually worth to a holder

The value of the capability is bounded by what the vote can reach. A vote that can move a treasury and upgrade contracts is a real claim on real decisions. A vote that can adjust a parameter nobody notices, while upgrade keys sit with a separate multisig, is close to worthless regardless of how it is marketed. Our governance entry covers the gap between formal and effective control and how to inventory it.

So the sizing question is concrete. Name what the timelock owns, name every key that can move funds or change code without passing through a vote, and the difference is what the feature is worth. Everything else is packaging.

Measured turnout, and who is actually deciding

The number people quote here is usually invented, so here is one that is not. Falk, Pathan, Rigas and Tsoukalas pulled onchain voting data across four DAOs and reported the average share of total token supply cast per proposal: 3.2% across 130 Aave proposals, 7.7% across 141 Compound proposals, 5.6% across 149 Lido proposals and 4.6% across 29 Uniswap proposals.³ Roughly 5% of tokens vote on a given proposal.

The second finding is sharper than the first. In the same dataset, for most votes the top three or four token holders together made up a majority of the votes cast.³ So the marginal holder is not buying a decisive vote. They are buying a right to be counted alongside a handful of addresses that decide the outcome, and any valuation of the feature has to start from that rather than from the theory.

None of this makes governance worthless. It makes it measurable, which is different, and it means the claim that a token has value because it governs something needs the same evidence as any other demand claim.

The design consequence, stated plainly

Attach the vote to a token that already has a reason to exist. Our utility token entry sets out the test we use for that: name the buyer, the day, the amount and the function that breaks without the token. Governance layered on top of a token that passes it adds a real right to something people already hold.

A token whose only concrete function is voting has to be valued on the vote alone, and that valuation runs straight into the turnout and concentration numbers above. In our experience across early stage design work this is the single most common structural mistake, and it usually arrives dressed as community ownership.

There is a classification dimension too, and it should be stated carefully. Where a buyer's return depends on decisions made by a team and a small set of large holders, the investment contract analysis has more to work with, and our Howey test entry walks those elements. Whether any specific token is a security is fact specific, jurisdiction specific and unsettled in places, and it is a question for counsel rather than for a glossary.

Five questions before adding the module

What does the timelock own. What can a passed proposal change without any other key being involved. What is the token's demand if every proposal stopped tomorrow. Is delegation live and used, or theoretical. And what would a buyer of a large position be able to force.

Answer those and the decision usually makes itself. The vote is infrastructure for a business that is already worth governing. It is not a substitute for having one, and nothing in this page recommends buying, selling or holding any token.