Back to research
Crypto & DeFi

Stablecoin Settlement for Banks: Cost, Regulation and Risk

A bank decision framework that separates ledger speed from end-to-end cost, legal finality, liquidity, custody and compliance.

Updated 8 min read
In this research

Stablecoins can move a token between blockchain addresses outside normal banking hours. That technical capability is relevant to banks, but it is not the same as saying every transfer is cheap, legally final or economically superior to an existing payment rail.

Visa announced USDC settlement for selected US issuer and acquirer partners in December 2025[1]. Separately, public-sector experiments are testing tokenised central-bank and commercial-bank money. Those developments show production use and continuing experimentation; they do not establish the best settlement asset for every bank, currency or customer.

Why this time is different

Regulation is now more legible in several markets, but it is not globally uniform. The EU's MiCA framework[2], the US GENIUS Act[3], Singapore's stablecoin framework and the UAE's payment-token regulation have different scopes, licensing routes and operating constraints. A token permitted in one corridor is not automatically permitted in another.

The right comparison is end to end. A blockchain network fee may be small and execution may be quick, but a bank must also price fiat conversion, liquidity and prefunding, custody, wallet controls, sanctions and transaction monitoring, investigations, reconciliation, redemption, capital treatment and exception handling. The BIS Committee on Payments and Market Infrastructures[4] cautions that potential cross-border benefits must be assessed alongside legal, operational and financial risks and divergent jurisdictional approaches.

Where the volume is actually coming from

Three candidate uses deserve testing: cross-border B2B corridors with poor speed or transparency; treasury movements between controlled group entities outside banking hours; and settlement between a network and participating financial institutions. "Candidate" matters. The originator and beneficiary may still need local fiat, which adds conversion and banking legs around the on-chain transfer.

A reproducible route-cost test

Compare one real corridor over a fixed observation period. For each rail, record quoted FX spread, explicit fees, liquidity or prefunding cost, compliance-review cost, failed or delayed payments, redemption time and the value date at the beneficiary's bank. Divide total cost by successfully delivered value. A low blockchain fee does not win the test if off-ramp, liquidity or exception costs are higher.

Where banks fit, and where they don't

A bank has more than one design choice. A third-party stablecoin creates a claim under the issuer's terms and applicable law. A tokenised deposit remains a commercial-bank liability represented on a programmable platform, as illustrated by JPMorgan's Kinexys[5]. Other options include custody, reserve banking, conversion or accepting an eligible token as one funding leg. The legal, balance-sheet and operational consequences differ, so the labels should not be treated as interchangeable.

The risks that have not gone away

Rules reduce some risks; they do not eliminate issuer, reserve, redemption, technology or intermediary risk. In March 2023, Circle disclosed that $3.3 billion of USDC reserves remained at Silicon Valley Bank[6] during a temporary depeg. A bank assessment should therefore examine the legal claim, reserve composition and location, redemption mechanism, concentration, chain and bridge exposure, key management, sanctions controls and recovery when a counterparty or network is unavailable.

Stablecoins, tokenised deposits, wholesale CBDC and retail CBDC, compared by issuer, balance-sheet treatment, primary use and maturity in 2026.
Stablecoin Tokenised deposit Wholesale CBDC Retail CBDC
Issuer Usually a non-bank issuer Commercial bank Central bank Central bank
Balance sheet Claim on issuer; reserve treatment depends on token and regime Stays on the bank’s balance sheet Central-bank liability Central-bank liability
Primary use Merchant, treasury, cross-border Institutional, programmable Interbank settlement Consumer payments
Maturity in 2026 Live, with use concentrated by token and market Live in selected institutional arrangements Pilots and limited live projects vary by jurisdiction Live, piloted or under review depending on jurisdiction

How CBDCs change the picture, and how they don't

CBDC, tokenised deposits and stablecoins are different liabilities and projects differ by jurisdiction. In May 2026, Project Agorá reported prototype results using tokenised commercial-bank deposits and central-bank reserves and said work would advance toward real-value testing[7]. That is evidence of a technical path, not proof of universal deployment or a reason to assume retail CBDC programmes have one shared status.

What a bank pilot has to prove

Before the first transaction, record the incumbent route's median and tail delivery time, total delivered cost, failure and repair rate, liquidity usage and compliance workload. Use the same definitions for the stablecoin route. Without a frozen baseline, a pilot can report fast blockchain confirmation while ignoring slower funding, redemption or investigation steps.

Four pilot tests for a bank stablecoin route, with the evidence to retain and an example stop condition for each.
TestEvidence to retainExample stop condition
Legal and eligibilityJurisdiction, token classification, contracts and redemption rightsRequired licence, approval or contractual right is absent
EconomicsFX, fees, liquidity, custody, compliance and exception costDelivered cost exceeds the agreed incumbent benchmark
OperationsConfirmation, reconciliation, failures, recovery time and manual workUnreconciled balance or recovery-time breach
Financial crimeWallet screening, alerts, disposition, escalation and audit trailCounterparty cannot be identified or an alert cannot be resolved

Volume alone is not a success measure. A useful pilot proves that the route works through exceptions, that balances remain inside approved limits, and that operations can return to the fallback rail without losing the transaction record or customer funds.

What a sensible strategy looks like

Start with a bounded corridor, named counterparties, an eligible token and chain, volume and balance limits, documented redemption, and manual fallback. Predefine stop conditions for depeg, liquidity, sanctions alerts, network disruption and reconciliation breaks. Only expand after comparing delivered cost, time, failures and operational losses with the incumbent rail.

For finance teams on the corporate side, the related accounting, custody and control questions are covered in our guide to stablecoins for treasury teams.

Evidence standard. Production-use claims use Visa's own announcement; regulatory links point to authorities; cross-border risk framing uses CPMI; and tokenised wholesale experimentation uses the BIS. Network fee and execution time are deliberately not presented as end-to-end price or legal-finality guarantees.

Sources

Numbered references are anchored to the specific claims they support. Primary documents are preferred wherever available.

  1. Visa announced USDC settlement for selected US issuer and acquirer partners in December 2025 investor.visa.com
  2. MiCA framework esma.europa.eu
  3. GENIUS Act whitehouse.gov
  4. BIS Committee on Payments and Market Infrastructures bis.org
  5. JPMorgan's Kinexys jpmorgan.com
  6. Circle disclosed that $3.3 billion of USDC reserves remained at Silicon Valley Bank circle.com
  7. Project Agorá reported prototype results using tokenised commercial-bank deposits and central-bank reserves and said work would advance toward real-value testing bis.org

Frequently asked questions

When can stablecoin settlement be cheaper than an existing rail?

Only when the complete route is cheaper after FX, liquidity, custody, compliance, redemption, reconciliation and exceptions, not merely when the blockchain fee is low. A bank should run a corridor-specific comparison using successfully delivered value and a common observation period.

Does on-chain confirmation create legal settlement finality?

It proves that a network accepted a state change under its protocol. Legal finality, discharge of the underlying obligation, redemption rights and treatment after a chain disruption depend on contracts and applicable law. Banks should document those points rather than infer them from confirmation speed.

How should a bank choose a stablecoin and chain?

Check legal eligibility in every relevant jurisdiction, the holder's claim, reserve and redemption terms, issuer and banking concentration, liquidity, chain and bridge risk, key controls, sanctions tooling and tested recovery. Then impose token, counterparty, volume and balance limits for the pilot.

What is the difference between a stablecoin and a tokenised deposit?

A stablecoin generally creates a claim under a non-bank issuer's arrangement, while a tokenised deposit represents a commercial-bank deposit on a programmable platform. Both can move on digital ledgers, but their issuer, balance-sheet treatment, protections and redemption path are different.

stablecoinspaymentsCBDC

The CloudFintech Briefing

Independent fintech analysis — AI in banking, payments, crypto, and regulation. No spam, unsubscribe any time.

By subscribing you agree to our Privacy Policy.