Stablecoin moats are usually judged by holder count. Is that the right unit of analysis?
Data and method
01Split stablecoin supply by who controls the decision to move it: individual holders, treasury allocators, platform governors, and protocol integrators.
02Tested the frame against four cases: USDT, Ethena's USDtb, Hyperliquid's USDH, and MakerDAO's USDC exposure, using DefiLlama history and governance records.
Findings
01
Decision points, not wallets
A hundred million retail wallets routed through three custodians are closer to three decision points. The proposed measure is decision-point concentration: a Herfindahl-style index with decision-makers, not addresses, as the unit.
02
A mandate is not liquidity
Hyperliquid's validators chose a native issuer for USDH, but supply stalled near $100M while USDC grew to about $5B. Circle kept the position by matching the yield-sharing terms, so its margin compressed even though nobody switched.
03
Switching is slower than it looks
MakerDAO governance rejected diversifying away from USDC in March 2023. USDC in the Peg Stability Module still fell from over $2B to under $500M by that August, through gradual substitution rather than a clean swap.
Limits
Holder counts for USDT come from Tether's own reporting, not an independent audit.
The closing prediction is meant to be falsifiable: a stablecoin distributed by governance mandate will lose share back to the incumbent within a year once the incumbent matches the economics.