Nostro reduction with stablecoins: how much a mid-size bank can actually retire

Nostro reduction with stablecoins is now a live treasury question rather than a research topic. Most mid-size banks and credit unions hold cash at foreign correspondents so payments can settle in local currency. That cash earns little and moves slowly. Settlement in regulated stablecoins completes in minutes on a public blockchain, so the logic behind pre-funding changes. The honest answer is that stablecoins can retire part of a nostro book, not all of it. This article sets out which balances can go, which must stay, and the controls a treasury team needs before it cuts anything.

What is a nostro account, and why does it tie up cash?

A nostro account is an account your institution holds at another bank, in that bank's currency. The mirror image, the account a foreign bank holds with you, is a vostro. These balances exist for one reason. A payment cannot settle in a currency you do not already hold somewhere.

The cost is rarely one line item. It appears in three places. Cash sits idle in low yielding accounts across several currencies. Each account carries its own reconciliation, statement handling and audit work. Each relationship also carries counterparty exposure to the bank holding the money.

The pressure is not new. The Bank for International Settlements Committee on Payments and Market Infrastructures has tracked a sustained decline in active correspondent banking relationships since 2011. Fewer correspondents means fewer places to hold currency. It also means less room to negotiate the balances they expect you to keep.

How much nostro can stablecoins actually replace?

Start with what the settlement leg does. It moves value between two parties in minutes, on a shared ledger, without a chain of intermediaries. Funding becomes on demand. You acquire the token when the payment is ready, send it, and the receiving side converts to local currency.

That removes the reason to hold cash in advance. Where a corridor can be served on demand, the standing balance behind it becomes optional. Nostro reduction with stablecoins is therefore corridor by corridor arithmetic. It is not a single switch.

The good candidates share a profile. They are outbound corridors with modest volume, unpredictable timing and a reliable local off ramp at the far end. Those accounts hold the most idle cash per payment made. They are also the accounts a correspondent is least interested in keeping open.

The realistic outcome is a smaller book, not an empty one. A team that maps balances by corridor usually finds a long tail of small accounts serving very few payments. That tail is where nostro reduction with stablecoins pays first, and where the internal case is easiest to defend.

Which nostro balances have to stay in place?

Some balances are not funding decisions at all. They are obligations. Identify them early and stop debating them.

A minimum balance written into a correspondent agreement stays until the agreement changes. Balances supporting intraday liquidity in a domestic real time gross settlement system stay. Currencies with capital controls, or with no compliant on ramp and off ramp, stay because there is nothing to replace the account with. Balances behind client sweeps and scheduled bulk payouts usually stay too, since predictable flows are exactly what pre-funding suits.

There is a supervisory dimension as well. Liquidity buffers under the Basel III liquidity coverage ratio framework depend on the quality and availability of the assets an institution holds. Moving working balances into a token changes the asset, the counterparty and the currency risk at once. That belongs in front of the risk committee and the supervisor, not in a treasury optimisation memo.

What does nostro reduction with stablecoins require operationally?

The mechanics are simple to describe and easy to underestimate. Every payment now carries a funding step, a settlement step and a conversion step. All three happen inside the same hour.

• Map the book first. List every nostro balance by currency, corridor, average balance, payment count and the reason it exists.
• Rank corridors by idle cash per payment. The worst ratio is the strongest candidate for an on demand settlement leg.
• Confirm the off ramp before anything else. A corridor only works if a licensed counterparty at the far end converts to local currency and pays out reliably.
• Run the corridor in parallel. Keep the nostro balance funded while a share of live volume moves on the new leg, then compare cost, timing and exceptions.
• Reduce in steps. Cut the balance to a floor, hold it for a full reporting cycle, and cut again only if the exception rate stays flat.
• Wire the accounting before go live. Decide how the token position is held, valued and reconciled, and who signs the daily position.
• Write the fallback. Name the trigger that reopens the balance, and keep the account open until that trigger has been tested.

The parallel run is the step most teams skip. It is also the only step that turns an estimate into evidence a risk committee will accept.

What controls belong in place before a balance is cut?

Cutting a nostro balance moves risk rather than removing it. Exposure to a correspondent becomes exposure to an issuer, a blockchain network and an off ramp partner. Each of those needs a named owner and a limit.

Issuer risk comes first. A regulated payment stablecoin is a claim on its issuer, backed by reserves. The GENIUS Act, signed into United States law in 2025, requires payment stablecoin issuers to hold reserves one for one in cash and short dated government securities and to publish monthly reserve reports. Read the attestations. Set an issuer limit the way you would set a bank line.

Network and operational risk come next. Fix the confirmation condition that counts as final on each chain you use. Whitelist destination addresses. Test the wrong network case deliberately, because sending on the wrong chain remains the most common way to lose funds outright.

Then compliance. The FATF Travel Rule applies to virtual asset transfers at or above a 1,000 USD or EUR threshold, so originator and beneficiary data must travel with the payment in IVMS101 form. Sanctions and PEP screening run on the parties. Transaction screening runs on the addresses. None of that is optional because the settlement leg got faster.

A nostro balance is not a cost you can simply delete. It is a risk you already chose and already priced. Retiring it means choosing a different set of risks, and pricing those with the same discipline.

How do you measure the benefit of nostro reduction?

Measure four things and ignore the rest. Each one belongs to a corridor, never to the programme as a whole.

First, released carry. Take the average balance retired in each currency and apply your internal funding rate. That is the annual carry the reduction returns.

Second, operational cost. Count the reconciliation hours, statement fees and maintenance charges attached to the accounts you closed or shrank.

Third, speed. Record the time from instruction to beneficiary credit on both legs, and the share of payments that needed manual repair. According to the Financial Stability Board's 2021 targets for the G20 cross border payments roadmap, 75 percent of wholesale cross border payments should be credited within one hour of initiation by the end of 2027. That is a fair external benchmark for either rail.

Fourth, the new costs. On chain settlement carries its own conversion spread, network fee and control overhead. Nostro reduction with stablecoins only counts as a saving once those are subtracted. Report the net result per corridor, because aggregate numbers hide the corridors that lose money.

One more discipline helps. Keep the payment record and the settlement record in the same format on both rails. If the on chain corridor produces a thinner audit trail than the correspondent corridor it replaced, the comparison is not fair and the examiner will say so.

Where StableNet fits

StableNet, built by SpendTheBits, is a cross border B2B payment and settlement platform for banks, credit unions, licensed money service businesses, exchange houses and remittance fintechs. Settlement is in regulated stablecoins on public blockchains, so a corridor can be funded on demand and completes in minutes with on chain auditability, which is the condition a nostro reduction depends on. The platform is ISO 20022 native, with pacs.008 customer credit transfers, pacs.009 interbank legs, pacs.002 status reports and pacs.004 returns inside head.001 envelopes, tracked end to end by UETR, so a corridor moved on chain keeps the same message record the treasury and audit teams already reconcile. Compliance is built in, with KYB and KYC onboarding, KYT, sanctions and PEP screening, FATF Travel Rule data in IVMS101 form, a compliance workbench and a tamper evident audit trail. SpendTheBits is a Bank of Canada registered payment service provider and a named finalist in the Swift Hackathon 2026 Technical Challenge.