A Tale of Two Payments
It is Monday morning. Two corporate treasury teams are each initiating a $500,000 cross-border payment from the United Kingdom to a supplier in the Philippines.
Treasury Team A goes through their bank. They submit a payment instruction. The bank routes it through a US correspondent, which routes it to a Philippines correspondent, which eventually delivers it to the beneficiary’s local bank. The money arrives on Wednesday. The net amount received is $499,215 — $785 lost in fees and exchange rate margins. The treasury team cannot track it in real time.
Treasury Team B uses a stablecoin payment service. They convert GBP to USDC through their bank’s licensed exchange. A blockchain transfer executes on Solana. The counterparty’s exchange converts USDC to PHP. The money is in the supplier’s account in under 60 seconds. The fee: under $1. (FX conversions at edges)
Same origin. Same destination. Same amount. Completely different experience.
This article is where stablecoins stop being theoretical and become practical. In Part 1, we covered what stablecoins are. In Part 2, we covered how they are created and destroyed. Now, in Part 3, we examine exactly how stablecoin payment flows work in the real world — and where they beat traditional rails, where they do not, and what the payments professional needs to understand about both.
First, Let Us Understand the Traditional Rails
Before comparing, you need a solid grasp of how traditional cross-border payments actually work. If you work in payments daily, some of this will be familiar — but the framing matters for the comparison.
(Please refer to other articles and youtube channel of PaymentTalks for more understanding on traditional flows)
The Correspondent Banking Model
When your bank sends a cross-border payment, it almost never sends it directly. It uses a chain of correspondent banks — banks that hold accounts with each other and act as intermediaries for international transfers.
This chain looks like this for a GBP-to-PHP payment:
Every hop in this chain:
- Charges a fee (typically $5–$20 per hop)
- Applies an exchange rate margin (typically 0.5%–2%)
- Adds processing time (each bank has cut-off times and processing windows)
- Reduces visibility for the originator
The SWIFT network carries the payment instructions (the messages) — but not the money itself. The actual funds flow through the bilateral correspondent account relationships. This is why SWIFT messages can travel fast while the actual funds settlement can take days.
What ISO 20022 / SWIFT MT Messages Do
In the SWIFT world, the payment instruction from Bank A to Bank B is carried in a standardised message:
- MT103: Single customer credit transfer (the traditional format)
- PACS.008: ISO 20022 equivalent — FI to FI customer credit transfer
- PAIN.001: The corporate-to-bank payment initiation message (upstream of the inter-bank transfer)
These messages carry structured data — Debtor, Creditor, Amount, Currency, Purpose, Remittance Information. The challenge is that as the payment hops through correspondents, some of this data gets truncated, reformatted, or lost entirely. This is one of the key drivers of the ISO 20022 migration — richer, structured data that survives the entire payment chain intact.
For stablecoin architects: this data richness challenge is equally present in stablecoin flows — and we will address it specifically in Part 5 of this series.
The Stablecoin Payment Flow — End to End
Now let us walk through the stablecoin equivalent of that same cross-border payment.
The Scenario
Debtor: ABC Manufacturing Ltd, Birmingham, UK. Holds a GBP account at Barclays. Creditor: Global Parts Inc, Manila, Philippines. Holds a PHP account at BDO Unibank. Amount: $500,000 USD equivalent.
The Flow: “Crypto Rails, Fiat at the Edges”
This is the most common enterprise stablecoin payment architecture — also called the on-ramp / off-ramp model. The client interacts with fiat at both ends. Stablecoins handle the middle.
Step 1 — ABC Manufacturing initiates the payment
ABC Manufacturing instructs their payment platform (or their bank’s stablecoin service) to send $500,000 to Global Parts’ wallet address. The instruction may look like a standard payment — Debtor, Creditor, amount, purpose. The stablecoin engine handles everything below.
Step 2 — On-Ramp: GBP converts to USDC
The payment platform converts GBP to USDC at the prevailing GBP/USD rate through a licensed exchange (e.g. Coinbase Prime, Bitstamp, B2C2). The exchange sells USDC from its liquidity pool and debits ABC Manufacturing’s account.
At this moment, ABC Manufacturing holds USDC worth approximately $500,000.
Step 3 — USDC transfer on Solana
The payment platform initiates a USDC transfer on the Solana blockchain from ABC Manufacturing’s wallet to the receiving exchange’s wallet in the Philippines. Solana validates and confirms the transaction. Settlement is final in approximately 400 milliseconds. The transaction is permanently recorded on the blockchain — visible to both parties and auditable.
Step 4 — Off-Ramp: USDC converts to PHP
The receiving exchange (e.g. Coins.ph, PDAX in the Philippines) receives the USDC. It sells the USDC from the pool and converts the USD value to PHP at the prevailing rate. The PHP is credited to Global Parts’ bank account at BDO Unibank.
Step 5 — Confirmation
Both parties receive confirmation. ABC Manufacturing can see the blockchain transaction hash. Global Parts sees the PHP credit in their account.
Total time: Under 60 seconds (excluding on/off-ramp FX settlement times, which vary by platform from seconds to minutes). Total cost: Exchange FX margin (typically 0.1–0.5%) + blockchain fee (<$0.01). No correspondent bank fees.
Three Major Use Cases in Detail
Use Case 1: Cross-Border B2B Payments — Replacing the Correspondent Chain
This is the highest-impact use case for large businesses. The correspondent banking chain is expensive, slow, and opaque. Stablecoin rails, when used as the settlement layer, can remove multiple hops, reduce fees by 90–98%, and deliver same-day (or same-minute) settlement.
Real-world example: Ripple’s ODL (On-Demand Liquidity)
Ripple’s payment network uses XRP (a cryptocurrency, not a stablecoin, but the mechanism is analogous) as the bridge currency in cross-border payment corridors. Rather than pre-funding nostro accounts in each currency corridor, the originating bank converts fiat to XRP, sends XRP across the network, and the receiving side converts XRP to local fiat — in seconds.
The same architecture works with USDC as the bridge asset. Companies like Bitso (Mexico), SBI Remit (Japan), and Tranglo (Southeast Asia) operate stablecoin-based payment corridors serving hundreds of millions of dollars in daily volume.
The practical implication for payments architects:
The correspondent banking model requires banks to hold pre-funded nostro accounts in each currency pair they support. This is capital-intensive. Stablecoin rails allow “just-in-time” liquidity — you convert fiat to stablecoin only when needed for a specific payment. This is a significant balance sheet efficiency improvement.
Use Case 2: Remittances — The Human Impact Case
The global remittance market is worth over $800 billion annually. It is one of the most important financial flows in the developing world — migrant workers sending money home to families who depend on it for food, education, and healthcare.
The World Bank estimates the global average cost of sending a remittance is approximately 6.3% of the amount sent. For some corridors — particularly in Africa and Southeast Asia — fees of 8–12% are not unusual.
The stablecoin remittance scenario:
A construction worker in Saudi Arabia, earning SAR 3,000/month, wants to send SAR 2,000 (~$533) to his family in the Philippines.
Traditional Western Union route:
- Fee: ~$25 (approximately 4.7%)
- Exchange rate margin: ~1.5%
- Time: Same day (if lucky), next day typically
- Net received by family: ~$495
Stablecoin route (via a licensed remittance app using USDC):
- Worker loads app, sends SAR to app’s local liquidity pool
- App converts SAR to USDC
- USDC transferred to Philippines partner exchange
- Partner converts USDC to PHP, delivers to family’s e-wallet or bank account
- Fee: ~$2–3 (approximately 0.5%)
- Time: Minutes
- Net received by family: ~$527
For a family surviving on remittances, that $32 difference per month is meaningful. Multiply that across the 200+ million migrant workers globally and the impact is measured in billions.
Companies actively operationalising this include Strike (using Lightning Network + stablecoins), Tempo (Europe-to-Africa corridors), and multiple players in the Philippines-US and Philippines-Middle East corridors.
Use Case 3: Corporate Treasury — Eliminating Trapped Liquidity
This is the use case that is generating the most interest at CFO and Treasury level in large multinationals.
The problem: A multinational corporation with subsidiaries in 15 countries holds working capital in 15 different local currencies at 15 different banks. Moving money between subsidiaries requires cross-border wires — fees, delays, FX conversion. The result is trapped liquidity — capital sitting idle in local accounts that cannot be efficiently deployed globally.
The stablecoin solution:
GlobalTech Corp has operations in the US, UK, Singapore, UAE, and Brazil. Instead of holding all working capital in local currencies, they implement a USDC treasury model:
- Each subsidiary converts idle local currency to USDC on a daily or weekly basis
- USDC is pooled in a central treasury wallet (or distributed wallets with aggregated view)
- When a subsidiary needs to pay a local supplier, the treasury releases USDC
- The subsidiary’s local exchange converts USDC to local currency and pays the supplier
Benefits:
- Netting: Global treasury can net off payables and receivables across subsidiaries before moving any money externally
- Speed: Intercompany transfers that took 2 days now take seconds
- Cost: Intercompany transfer fees eliminated entirely
- FX optimisation: Convert from USDC to local currency at the time and rate that suits the treasury
Companies like Coinbase, Circle, and Fireblocks have published case studies of Fortune 500 companies using this model. PayPal uses its own stablecoin (PYUSD) for internal treasury purposes.
The Head-to-Head Comparison
Here is the complete picture for payments architects making a rail selection decision:
| Dimension | SWIFT (MT103 / PACS.008) | SEPA Credit Transfer | Stablecoin Rails |
| Settlement Time | 1–3 business days | D+1 (next business day) | Seconds to minutes |
| Cross-Border Fees | $25–$75+ per transaction | < €0.50 (SEPA only) | < $1 (often <$0.01) |
| Operating Hours | Business hours, weekdays | Business hours, weekdays | 24/7/365 |
| Currency Coverage | 140+ currencies | Euro zone only | USD/EUR dominant, growing |
| Reversibility | Possible (recalls, R-txns) | Possible (8-week window) | Generally irreversible |
| Data Richness | ISO 20022 (rich) / MT (limited) | ISO 20022 (rich) | Minimal native metadata |
| Regulatory Clarity | Fully established | Fully established | Evolving rapidly |
| AML/KYC Framework | Mature (FATF, CBPR+) | Mature (EPC) | Maturing (FATF VASP rules) |
| Dispute Resolution | Established process | Established process | Limited / manual |
| Network Reach | 11,000+ institutions | 36 SEPA countries | Growing institutional adoption |
| Transparency | Limited (correspondent hops) | Limited | Full on-chain visibility |
| Minimum Viable Amount | Typically $1,000+ economical | Any amount | Any amount, even micro-payments |
Where Stablecoin Rails Win — and Where They Do Not
Being intellectually honest about this is essential for architects. Recommending the trendy solution over the right one serves no one.
Where Stablecoins Have a Clear Advantage
High-fee cross-border corridors.
Anywhere SWIFT correspondent banking charges $30–$75+ per transaction, stablecoin rails at $0.01 are transformational — especially for mid-market businesses making frequent payments.
24/7 settlement requirements.
SWIFT has cut-off times. SEPA processes overnight. A supplier in a different time zone who needs payment at 2am on a Sunday cannot be served by traditional rails. Stablecoins have no cut-off times.
High-volume, low-value payments.
Sending $5 via SWIFT is economically nonsensical. On Solana, it costs a fraction of a cent. This enables entirely new payment models — micro-payments for content, per-use billing, streaming payments.
Emerging market corridors.
The Philippines, Nigeria, Kenya, Indonesia, and similar markets are historically expensive and unreliable for SWIFT transfers. Stablecoin corridors have emerged specifically to serve these flows.
Intra-company treasury movements.
There is no customer protection argument for routing intra company transfers through correspondent banks. Speed and cost efficiency win here unambiguously.
Where Traditional Rails Remain Stronger
Reversibility and dispute resolution.
A SWIFT payment can be recalled (with effort). A SEPA payment can be returned within 8 weeks. A blockchain transaction is final. For consumer payments, B2C e-commerce, or any scenario where fraud recall is necessary, traditional rails provide protections that stablecoin rails cannot currently match.
Domestic retail payments.
Within a country, SEPA Instant, UK Faster Payments, Indian IMPS, and similar domestic real-time payment systems offer speed comparable to stablecoins — with full regulatory protection, reversibility, and consumer trust. Stablecoins offer no significant advantage here.
High-value, regulated settlements.
For transactions above $10 million, Fedwire (US) and CHAPS (UK) provide same-day finality with full central bank backstop. The liquidity depth of stablecoin exchanges for single transactions of this size can be a limiting factor.
Jurisdictions with unclear or hostile crypto regulation.
China, India (evolving), and some GCC countries restrict or prohibit stablecoin payments for domestic use. A payment architecture that relies on stablecoin rails in these jurisdictions may face regulatory challenge.
The Architect’s Mental Model: Rail Selection Framework
When assessing whether stablecoin rails are appropriate for a given payment flow, use this structured evaluation:
Step 1 — Assess the corridor
Is the corridor expensive (>$10 in fees)? Is it slow (>1 day)? Are traditional rails unreliable or unavailable? If yes to any of these → stablecoin rails are worth evaluating.
Step 2 — Assess the volume and value
High frequency, low value? Strong case for stablecoins. Very high value (>$5M single transaction)? Assess exchange liquidity carefully.
Step 3 — Assess reversibility requirements
Does the payment scenario require recall capability, fraud dispute resolution, or regulatory consumer protection? If yes → traditional rails are required or a hybrid model is needed.
Step 4 — Assess the regulatory environment
Is stablecoin payment activity licensed and compliant in both the originating and receiving jurisdiction? Do both parties have the necessary VASP registration? If not → do not proceed until this is resolved.
Step 5 — Assess the counterparty
Does the receiving party have a stablecoin wallet (or access to an off-ramp)? Is their exchange reliably converting to their local currency? The off-ramp is frequently the weakest link in the chain.
Common Misconceptions About Stablecoin Payment Flows
“Stablecoins eliminate FX entirely.”
No. Stablecoins eliminate the FX conversion friction within the payment chain — the multiple conversions through correspondent relationships. But if you are paying a supplier in Philippine Pesos, someone still needs to convert from USD (stablecoin) to PHP. The FX conversion simply happens at a more transparent and competitive point — the off-ramp exchange — rather than embedded in opaque correspondent margins.
“Stablecoin rails mean I can ignore SWIFT entirely.”
For now, no. The vast majority of global payment volume still flows through SWIFT. Many stablecoin payment solutions operate alongside SWIFT — they use SWIFT for the data messaging layer and stablecoins for the settlement layer. SWIFT itself is actively piloting tokenised asset settlement. The two worlds are converging, not competing.
“Blockchain transparency means my payment details are public.”
Partially true. The blockchain records wallet addresses and amounts publicly — but not the identity behind the wallet address. On a public blockchain like Ethereum, anyone can see that wallet address 0x742d… sent 500,000 USDC to wallet 0x9b3c…. They cannot see that those wallets belong to ABC Manufacturing and Global Parts unless that information is published separately. For institutional payments, this is a compliance design consideration — on-chain analytics firms can often de-anonymise large flows through exchange data.
Key Takeaways
- Stablecoin payment rails follow a “fiat at the edges, stablecoin in the middle” architecture. Clients deal in local currency. The stablecoin handles the cross-border settlement layer.
- The on-ramp (fiat → stablecoin) and off-ramp (stablecoin → fiat) are the critical integration points. Their fee structure, liquidity depth, and regulatory status are as important as the blockchain transfer itself.
- Stablecoins have a decisive advantage in cross-border corridors that are expensive, slow, or underserved — particularly emerging market corridors and 24/7 settlement requirements.
- Traditional rails retain advantages in reversibility, consumer protection, high-value settlements, and jurisdictions with clear regulatory frameworks for traditional banking.
- The smart architect designs for both rails — using stablecoins where they excel and traditional infrastructure where regulations, reversibility, and risk management require it.
What’s Next
You can now trace a stablecoin payment from originating corporate to beneficiary bank — and you know when to recommend stablecoin rails over SWIFT and vice versa.
But none of this works without understanding the regulatory and risk landscape. Before you recommend stablecoins to a client or include them in a payment architecture, you need to know the six risks that can break the model — and the three major regulatory frameworks that are reshaping what is legally permissible.
Part 4 of this series is the one I would make mandatory reading before any architect touches a production stablecoin system.
