Key findings
- Conversion, funding, redemption and exceptions can matter more than network fees.
- Hold amount, currencies, receiving method and time conditions constant in comparisons.
- A token-to-token transfer and a fiat-to-bank payment are different services.
Where does the payment start?
Comparing a stablecoin network fee with an all-in bank remittance price compares different tasks. A transfer between two wallets that already hold tokens is different from starting with Korean won and delivering usable cash to an overseas bank account. Fix the route before explaining the cost difference.
The World Bank’s remittance-price methodology considers exchange-rate margins and service conditions alongside explicit fees. Its approach is useful, but a retail remittance average is not a matched quote for a corporate payment or token transfer. [1]
| Component | Assumed cost |
|---|---|
| Entry conversion: 0.4% | $4.00 |
| Network transfer | $0.10 |
| Exit conversion and payout: 0.6% | $6.00 |
| Allocated support and exceptions | $1.00 |
| Total | $11.10 / 1.11% |
Illustrative assumptions, not provider quotes or market averages. Funding cost is assumed to be zero.
Speed is both a cost and a promise
A fast on-chain transfer does not complete a payment if the recipient still waits for fiat. If the recipient can use tokens directly, the exit stage may disappear. Combining these cases in one average processing time conceals the most consequential distinction.
A provider can prefund a payout so that customers receive funds sooner. That consumes capital even when its cost is absent from a displayed transaction fee. Identify who finances the promise of immediate availability.
Exceptions determine scalability
Wrong networks, recipient restrictions and delayed customer checks can require manual work. A small exception rate can dominate unit economics when each case takes a long time. Success-only transaction fees do not capture those costs.
Direct-redemption terms constrain the available exit route. A provider’s institutional access need not be available to its end customers. Customer promises must describe the route customers actually have. [2]
| Dimension | Hold constant |
|---|---|
| Amount | Payment size and minimum/maximum limits |
| Currency | Funding, receipt and exchange-rate reference |
| Receipt | Wallet balance or bank credit |
| Timing | Business days, weekends and promised deadline |
| Exceptions | Rejection, refund and error-correction costs |
CBK analysis. This is an analytical framework, not observed customer performance or a market benchmark.
Where the advantage can grow
When both parties retain the same settlement asset and transact repeatedly, entry and exit costs can be spread over several tasks. Fragmented existing payment routes and time-zone gaps can also increase the value of an alternative. Legal access and usable cash-out routes still have to work.
For a one-off small payment requiring conversion and onboarding at both ends, low network fees alone may offer little advantage. Start with the expensive stage of the customer journey. BIS analysis helps frame settlement and trust as connected functions rather than isolated technical features. [3]
The cheapest route changes with payment size
Suppose route A costs $5 plus 0.2% and route B costs $0.10 plus 0.8%. Their costs are equal at approximately $816.67. Below that amount B benefits from its lower fixed charge; above it A benefits from its lower proportional charge.
Real quotes include minimum charges, caps, exchange-rate changes and route restrictions. This linear model is not a transaction quote. It helps choose pilot sizes: test small and large payments rather than only the average ticket.
| Hypothetical payment | A: $5 + 0.2% | B: $0.10 + 0.8% |
|---|---|---|
| $100 | $5.20 | $0.90 |
| $1,000 | $7.00 | $8.10 |
| $10,000 | $25.00 | $80.10 |
CBK analysis. This is an analytical framework, not observed customer performance or a market benchmark.
Run a small pilot with an end-to-end unit
Link completion, actual received amount, elapsed time and exception-handling time to one transaction identifier. Record the difference between quoted and realized cost. This distinguishes network-price changes, exchange margins and human intervention.
A handful of successful transactions cannot establish service-wide economics. Report the observed routes and leave untested hours and recipient conditions explicit. Transparent costs can be more durable than an unsupported headline saving.
Scope & limitations
No live payment or remittance quotes were collected. Hypothetical costs explain structure and do not guarantee savings. Country-specific regulatory, tax and sanctions conditions require separate review.
CryptoBridge Korea analysis based on public materials. Source facts are distinguished from interpretation and hypothetical calculations. References to organizations or projects do not imply their sponsorship or review of this paper.
Sources & scope
- World Bank, Remittance Prices Worldwide methodology ↗Remittance fees and exchange-rate margins. Not directly comparable to corporate payment costs.
- Circle, USDC Terms ↗Distinguishes holding a token, direct-redemption eligibility and contractual restrictions.
- BIS, Annual Economic Report 2025, Chapter III ↗BIS analysis and policy perspective on monetary and payment structures.