What the fee-generating assets look like on 22 October 2026
A cross-asset look at annualized fee streams and market-cap-to-fee multiples as of 22 October 2026, with notes on what the figures can and cannot support.
The Spread of Market-Cap-to-Fee Multiples
As of 22 October 2026, the market-cap-to-fees ratio across the fourteen assets in this dataset ranges from 3.5x (Uniswap, annualized from the trailing 30 days) to 18,821x (Bitcoin, annualized from the trailing 30 days). That is a spread of roughly four orders of magnitude. A wide spread of this kind tells a reader that fee generation is distributed very unevenly relative to market capitalization across the group; it does not, by itself, tell a reader anything about whether any individual multiple is appropriate, excessive, or deficient. The ratio is a description of a relationship between two figures at a point in time, not a signal about direction or value.
Fee Retention: Take-Rate Divergence
Several assets in the dataset retain all or nearly all of the fees their networks collect. TRON, Hyperliquid, Canton, and Avalanche each show a take rate of 100% (annualized from the trailing 30 days), meaning revenue equals fees in the data as reported. Chainlink's take rate stands at 97.8% (annualized from the trailing 30 days). At the other end, Uniswap retains approximately 8.1% of its $1.26 billion in annualized fees (annualized from the trailing 30 days), passing the remainder to liquidity providers. Cardano retains roughly 20.0% of its fees, and BNB retains 10.0%. Where revenue is null—as it is for Bitcoin and Zcash—the dataset does not supply a split, so no take-rate comparison can be made for those assets. A high fee figure paired with a low take rate means the network or protocol intermediates a large volume of economic activity while capturing a smaller share of it as protocol-level income.
Chain-Basis vs. Application-Basis Figures Are Not Interchangeable
The accounting basis column in this dataset distinguishes between chain-level figures and application-level figures. Uniswap's fees are sourced from application:Uniswap V3; Chainlink's from application:CCIP. Every other asset draws from a chain-level source. Chain-level fees aggregate all activity settled on that ledger; application-level fees measure only what flows through a specific protocol deployed on top of some underlying chain. Comparing Uniswap's market-cap-to-fee ratio directly against Ethereum's, for example, conflates two different scopes of measurement. The application figure is a subset of broader on-chain activity and does not carry the same economic perimeter as a chain-level number. Readers should treat the two subgroups separately rather than ranking them in a single undifferentiated table.
Why These Are Not Equity Multiples
The single most important structural difference between a market-cap-to-fee ratio and an equity price-to-revenue or price-to-earnings multiple is that token holders generally do not hold a legal, enforceable claim on fee cash flows. In an equity context, revenue flows through a corporate structure to shareholders under defined legal rights. Token economics vary by protocol design, governance, and jurisdiction; the relationship between fee generation and any return to token holders is contingent, not contractual. Treating a market-cap-to-fee ratio as equivalent to a price-to-sales multiple therefore imports an assumption about cash-flow rights that the data does not support.