A Digital Dollar Still Needs a Door to the Bank

USDC can move between wallets. Turning it into bank dollars, and receiving income from the assets behind it, depends on a different set of relationships.

Imagine acquiring 100 native USDC at par, before any exchange, bank or network charges, and then sending the tokens to another wallet. This is an illustration, not a transaction we made or a quote a customer can execute. The recipient can hold or transfer the tokens on a supported network, subject to applicable transfer restrictions. But if that recipient wants dollars in a bank account, possession of the tokens is only the beginning of the story. The route depends on where the holder is, whether the holder has an issuer account, and what checks the issuer requires. Circle’s non-EEA USDC terms, §§1–2; Circle France’s EEA redemption policy, §§1–4.

The phrase “digital dollar” covers three different things. A token can be transferred. Its issuer holds dollar-denominated assets to back it. A person can have a route to turn it into bank money. Those features sit together in USDC, but they do not confer identical rights on every holder. Nor does the interest earned on reserve assets automatically pass to the person holding the token. The issuer’s policies and its distributor agreements decide much of what happens after the transfer.

Three ways out of 100 USDC

Consider first an eligible institution outside the European Economic Area (EEA) with a Circle Mint account in good standing. Under Circle’s USDC terms, updated 12 December 2025, a Mint customer can redeem USDC for dollars directly with Circle, subject to the terms. Circle’s Mint agreement, updated 28 September 2026, §1 limits the service to institutions in supported jurisdictions. Registration requires identity information and Circle may refuse an account. A direct issuer route exists for an institution that passes onboarding and keeps its account in good standing. These terms alone establish neither acceptance of a particular applicant nor the settlement time or all-in cost of a real request.

Now send the same 100 USDC to a holder outside the EEA who has no Mint account. Circle’s non-EEA terms call this a holder without an account, rather than a Circle customer. Section 2 says such a holder may not redeem directly with Circle unless and until they open a Mint account. Since Mint is currently restricted to institutions in supported jurisdictions, owning the token alone does not open that door. The holder may instead seek another person or venue willing to take the token in exchange for bank dollars, subject to that venue’s terms and price. An intermediary might offer an exit, but these terms establish neither its availability to this holder nor its price.

The third holder is an eligible retail person in the EEA. Circle France’s policy, updated 15 September 2026, §§1.3–2.3 and 4.2 offers a direct form-based redemption route for retail holders who are not Circle Mint customers. The holder can request redemption from Circle France without first being refused by an exchange. Circle France assesses eligibility, checks identity and transaction history, and requires an EEA bank account that can accept USD. The holder receives a deposit address for the tokens and, after the required checks, payment goes to the bank account. The policy says the ordinary process should take no more than five business days unless discrepancies arise. Section 8.4 allows adjustments during a defined reserve-rebalancing stress event. This is the issuer’s stated process, not evidence that our hypothetical holder passed the checks or received dollars in five days.

Scroll the table sideways to read all columns.

The table maps policies, not a realised 100-USDC cash flow. The starting assumption of par acquisition excludes unspecified charges; an intermediary’s price could differ from par. Circle France also warns that third-party bank charges may apply to USD payments. No verified fee schedule in this evidence packet supports an all-in exit figure. Circle’s non-EEA terms describe powers to block transfers to certain addresses and freeze associated assets in specified circumstances, including legal orders. The terms establish a control the issuer says it has, not an observed freeze or a smart-contract audit. USDC terms, §13; EEA policy, §§4.2 and 8.4.

The comparison holds token quantity fixed, so a different route cannot be attributed to a different-sized balance. Circle’s non-EEA direct route depends on an approved institutional Mint account; Circle France’s retail route depends on its EEA policy and checks. The non-EEA holder without Mint gains neither route simply by receiving the tokens. A wallet transfer changes who holds them, but it does not complete issuer onboarding, identity checks or a bank payment. That separation, rather than a measured difference in fees or speed, is what the policies establish.

The income behind the token

Circle says each USDC is backed by a dollar or equivalent dollar-denominated assets. Those assets can earn income. Under its non-EEA USDC terms, §2, holding USDC does not itself entitle a holder to interest or other returns on the reserves. That does not mean a holder can never receive a reward. An exchange can offer one under a separate programme, with its own eligibility and risks. The BIS’s June 2026 study of stablecoin remuneration already distinguishes rewards funded from reserve income from those linked to exchange activity. Revenue sharing is therefore established prior work, not a discovery of this article.

The narrower question here is how a token’s ability to travel relates to the issuer’s reported economics. Circle’s Q2 2026 Form 10-Q reports income and costs for the three months ended 30 June 2026 alongside the same quarter of 2025. The numbers are for the consolidated company. Reserve assets backing both USDC and EURC generate the reported reserve income, so the table cannot assign every dollar to the holder of our hypothetical USDC.

Scroll the table sideways to read all columns.

Source: Circle’s 10-Q Condensed Consolidated Statements of Operations, printed p. 8 (original HTML table 16, zero-based); distribution-cost disclosure, p. 47; and quarterly management comparison, pp. 49–50. Statement cells were reported in USD thousands and converted here to millions. The Coinbase row is a separately disclosed rounded component of distribution and transaction costs, not an additional expense to add to the total. The net-income row includes items beyond reserve income and distribution.

Reserve income rose by $33.459 million, or 5.28%, between the matched quarters. The separately reported Coinbase distribution cost fell by $7.7 million, or 2.32%. In these matched quarters, a larger reserve-income pool coincided with a smaller Coinbase expense. Circle attributes that decline to a decrease in Coinbase’s share of total average USDC on-platform balances. That is management’s explanation; these two totals alone do not identify a causal effect.

The Coinbase cost was about 79.09% of the $410.414 million distribution-and-transaction-cost category in Q2 2026, compared with 81.75% of the $406.472 million category a year earlier. The numerator is rounded and the denominator includes other distributors and network transaction costs. These ratios show the size of one disclosed expense within a matched category. They are not Coinbase’s contractual percentage of reserve income, a customer payout, or economic rent.

The filed arrangement gives more context than a simple “Circle pays Coinbase half” description. Circle’s Q2 10-Q says Coinbase receives allocations associated with USDC held on its platform after issuer retention. It also receives half of the remaining amount tied to broader ecosystem growth after payments to approved third-party participants. The 18 August 2023 collaboration agreement, Exhibit 10.1, §§3.1–3.2 and 4.1–4.5 has redacted terms and schedules. Neither document supplies a complete calculation for one token, and the filing does not say Coinbase takes half of gross reserve income. The historical allocation is informative without establishing the agreement’s present renewal status.

A later Circle filing dated 22 September 2026 shows why the quarter and contract date matter. It reports new Binance arrangements entered into on 17 September that replace earlier ones and include a monthly incentive tied to USDC held through Modular Smart Contract Wallet services. The disclosed summary does not state the percentage. These terms came after the June quarter and cannot be read back into its cost table. They do show that distribution economics can change while the token remains transferable.

The strongest case for distributors is operational. They can make USDC easier to acquire or sell, offer a liquid venue, carry out compliance checks and sometimes pay rewards. A holder without Circle Mint access may value those services. The EEA route also shows that direct retail issuer access exists under a different policy. Yet none of the reported income and expense totals tells us what a given holder paid, earned or would have preferred. The 2025 loss-to-2026 profit movement cannot be credited solely to reserve or distribution growth without examining the other expenses and items behind net income.

The 100-USDC thought experiment ends where the balance sheet cannot: at the holder’s exit choice. The token can change wallets, but bank dollars arrive through an issuer or intermediary with its own eligibility, checks, timing and price. Reserve income follows still another path through company accounts and distribution agreements. To judge the arrangement for a real holder, we would need that holder’s permitted route, actual price and charges, settlement record and any separate reward terms. The published documents show the doors and some of the income flowing around them; they do not show which door our hypothetical holder could open at what net value.

Evidence cutoff: 30 September 2026 UTC (1 October 2026 IST). Policies and filings are described at their stated dates.