What USDC-denominated gas means for a builder's budget

Arc denominates gas in USDC rather than a volatile native token — the mechanics of that (EIP-1559-style base fee, 18-decimal accounting, the paymaster system) belong to Circle's own documentation at docs.arc.io. What this guide covers is the budgeting consequence: a dollar-denominated fee lets a team forecast infrastructure cost in its own reporting currency instead of hedging a token price, gas becomes a line item rather than a treasury position, and a wallet holding zero USDC simply cannot transact — which is a real operational constraint, not a theoretical one. The actual per-transaction USDC cost at mainnet is not yet confirmed as of 2 September 2026, and this guide does not estimate it.

The mechanics live at docs.arc.io — this is the budgeting angle

For how Arc's fee model actually works — base fee, priority fee, the EIP-1559-style adjustment mechanism, and the paymaster system that can let other stablecoins cover gas indirectly — see docs.arc.io. Built on Arc's job here is narrower: what does gas being USDC-denominated, specifically, change about how a team plans and reports infrastructure spend, and what should a technical decision-maker actually budget for going into mainnet.

Forecastable in your own reporting currency

On a chain where gas is paid in a volatile native token, a team's infrastructure cost forecast is really two forecasts multiplied together: expected transaction volume, and the token's price over the budgeting period. The second variable is outside the team's control and can swing the actual dollar cost of a fixed transaction volume by a wide margin between the forecast and the spend. Gas denominated in USDC collapses that to one variable — transaction volume — because the unit the fee is paid in and the unit the budget is reported in are the same. A finance function forecasting infrastructure spend for Arc in dollars is not translating a token-price assumption into that forecast; it's translating gas usage, full stop.

An opex line instead of a treasury position

The practical consequence for how a team manages funds: gas budgeting on Arc looks more like paying a cloud-hosting bill than managing a crypto treasury position. There's no separate task of acquiring and holding a native token purely to keep transactions moving, no exposure to that token's price between funding the wallet and spending the gas, and no mark-to-market swing on the gas reserve itself between accounting periods. USDC held for gas is, from an accounting perspective, functionally the same as USDC held for anything else the business does — one balance, one unit, no separate volatility bucket to explain in a board deck.

The constraint this doesn't remove: funding and monitoring gas float

None of the above removes the operational work of keeping gas-paying wallets funded. A hot wallet with zero USDC in it cannot transact on Arc at all — there's no fallback to a different native token, because USDC is the only gas asset the protocol accepts (outside whatever a paymaster arrangement covers on the user's behalf). That means every wallet that needs to initiate transactions autonomously — a backend service signing on a schedule, a bot, an automated settlement flow — needs its USDC gas float actively monitored and topped up before it hits zero, not after. For a team running several such wallets across environments, that's a real operational surface: alerting on balance thresholds, a process for replenishing float, and a decision about who's authorized to move funds into gas wallets. It is a smaller, more predictable version of the same problem as running out of a cloud budget mid-month, not a new category of risk, but it doesn't go away just because the unit is a stablecoin.

What this guide will not do: estimate the mainnet cost

Circle's testnet documentation states a base-fee *target* of roughly $0.01 per transaction under normal load, explicitly labeled as current testnet configuration, with the accompanying note that the underlying fee parameters — minimum and maximum base fee, throughput limits — may change before mainnet. Built on Arc will not carry that testnet target forward as a mainnet estimate, and no official mainnet fee figure has been published as of 2 Sep 2026. A team building a per-transaction cost model for board or investor reporting should treat that number as unconfirmed until Circle publishes mainnet parameters, not plug in the testnet target and call it a forecast.

Earning on idle gas float isn't this guide's question

A natural follow-up question — can a team earn yield on the USDC it holds as gas float rather than letting it sit idle — is a real treasury-management question, but it's an earn/yield question, not a gas-budgeting one, and it belongs to ArcYield, not Built on Arc.

Sources: - Arc gas and fees reference - Arc developer documentation - Arc EVM differences reference

*Built on Arc is an independent directory. Arc is a Circle product; we are not affiliated with, endorsed by, or operated by Circle.*