Intraday liquidity buffer for a stablecoin corridor: how to size it and when to move it
A corridor that never closes needs a buffer sized on peaks, not daily averages. How to size an intraday liquidity buffer and the triggers that move it.
- An intraday liquidity buffer is sized on the peak, not the daily average. The Basel Committee on Banking Supervision's April 2013 standard, Monitoring tools for intraday liquidity management, requires banks to report daily maximum intraday liquidity usage precisely because an average hides that peak.
- A stablecoin corridor has two legs on two clocks. The on chain settlement leg has no cut off and no holiday, the fiat payout leg does, and the buffer exists to absorb the mismatch.
- Size each leg separately from ninety days of your own timestamped payment data, take the worst day rather than the mean, and hold the buffer for the funding time you actually measured.
- Write the triggers before you need them: a volume band, a client concentration limit, a slowdown in the funding path, and a risk event such as a depeg indicator or a chain outage. Each one needs a named owner and a time limit.
- The weekend is the real exposure. Payments keep arriving while fiat funding is shut, so stress a Friday evening to Monday morning window on its own.
An intraday liquidity buffer is the working float a corridor holds so payments clear without waiting for funding. In a stablecoin corridor that question gets harder, not easier. Settlement completes in minutes and the settlement leg never closes, so there is no overnight window in which positions quietly reset. A treasury team that sizes an intraday liquidity buffer for a stablecoin corridor from a once a day balance report holds too much float most of the week and too little on the two days that matter. This is how to size the buffer, where to hold it, and which triggers should move it.
What is an intraday liquidity buffer in a payment corridor?
It is the balance you hold in each leg of the corridor to meet payment obligations during the business day rather than at the end of it. Two legs, two buffers. There is the stablecoin float you hold to settle out, and the local currency float your payout partner holds to deliver funds to the beneficiary.
The distinction matters because an end of day position can look healthy while the intraday position fails. A corridor that nets flat over a day can still run dry at the hour its payments cluster. Supervisors made this point long before stablecoins existed. The Basel Committee on Banking Supervision's April 2013 standard, Monitoring tools for intraday liquidity management, requires banks to report their daily maximum intraday liquidity usage and the available intraday liquidity at the start of each business day, because a daily average hides the peak.
The buffer is not a capital requirement. It is an operational decision you own, and it has a price. Every unit of float parked in a corridor is a unit earning nothing elsewhere.
Why does a settlement leg that never closes change the buffer?
Because the two sides of the corridor now run on different clocks.
A public blockchain has no cut off, no business day and no currency holiday. A stablecoin leg settles in minutes on a Sunday. The payout leg does not. Local clearing systems keep their own hours, and the beneficiary bank keeps its own. The buffer absorbs that mismatch.
That reshapes the problem in three ways. Peaks arrive at any hour, so a position checked each morning is blind for most of the day. Funding is faster, which argues for a smaller buffer, but only if you can really move value when you need it. And the weekend becomes a live exposure rather than a quiet period, because payments keep arriving while the fiat side is shut.
The wider market is moving the same way. The Financial Stability Board's October 2021 targets for addressing the four challenges of cross border payments require that, by the end of 2027, 75 percent of cross border wholesale payments are credited to the beneficiary within one hour of initiation. One hour leaves no room to arrange funding after the fact.
How do you size an intraday liquidity buffer for a stablecoin corridor?
Size it from your own payment data, not from a rule of thumb. The method is the one supervisors already use for intraday liquidity. Find the peak, not the average.
- Pull at least ninety days of corridor payment data, timestamped in one time zone. Value, volume, hour of arrival and payout completion time.
- Build the intraday profile. Chart cumulative outflow against cumulative inflow by hour. The widest gap between the two curves is your peak net requirement, and that is the number the buffer has to cover.
- Separate the two legs. Size the stablecoin float against the settlement obligation and the local currency float against the payout obligation. They peak at different hours.
- Take the tail, not the mean. Use the worst day in the window, then add a margin for the single largest payment one client can send.
- Measure the funding path. Time how long a top up actually takes in each leg, from decision to usable balance, including screening and any partner cut off.
- Hold the buffer for the funding time, not for the calendar day. If a top up lands in thirty minutes, the buffer covers the peak for thirty minutes. If it takes two days, it covers two days.
- Stress the weekend and the holiday on their own. Run the profile across a Friday evening to Monday morning window with no fiat funding available.
- Write the number down with its date, its inputs and its owner, then review it on a stated cadence.
The output is not one number. It is a floor, a target and a ceiling for each leg, plus the trigger that moves you between them.
A buffer sized from an end of day balance is sized for a problem you do not have. The question is never what the corridor held at midnight. It is what it held in the hour its payments clustered.
What triggers should move the buffer up or down?
Set the triggers in advance and in writing, because the moment you need one is the moment nobody wants to make a judgement call.
Four kinds cover most corridors. A volume trigger moves the buffer when rolling daily value crosses a stated band. A concentration trigger moves it when one client's share of corridor value crosses a limit, because a single large sender changes the peak. A timing trigger moves it when the funding path slows, for instance when a partner changes a cut off. A risk trigger moves it on an event, such as a depeg indicator, a chain outage or a sanctions hold that strands value.
Each trigger needs a named owner, a stated action and a time limit. Who may release funding, how much, and by when. A trigger with no named owner is a paragraph, not a control.
Keep the review cadence separate from the triggers. Triggers handle the exception. A monthly review handles the drift, because corridor volume rarely moves in one step.
Where should the buffer sit, and in what asset?
Split it deliberately. Three places, three purposes.
The settlement float sits in the stablecoin you actually settle in, in custody you control, ready to move. The payout float sits in local currency with the partner who delivers funds, governed by a written agreement on who owns it and what happens if that partner fails. The reserve sits outside the corridor in something liquid, and it funds the other two when a trigger fires.
The asset choice is a credit decision, not a convenience. Read the issuer attestation reports and check reserve composition, attestation cadence and redemption mechanics before you hold a float in any token. A float you cannot redeem on demand is not liquidity.
Scale the reserve to the funding time you measured, and keep it out of the corridor. The Basel Committee's 2013 Liquidity Coverage Ratio standard requires banks to hold unencumbered high quality liquid assets sufficient to survive a thirty calendar day stress scenario. The principle carries down to a single corridor. The buffer that funds a buffer has to be available without selling into a stressed market.
What usually goes wrong with intraday liquidity in a corridor?
The failures are operational, and they repeat across institutions.
Sizing from the end of day net position. The corridor nets flat, the treasurer concludes it needs almost nothing, and then a Tuesday peak forces a manual top up at the worst moment.
Counting a pending balance as an available one. Confirmation depth is a risk decision. Value counted before you would accept it as final is not in the buffer.
One buffer for both legs. The stablecoin side and the local currency side peak at different hours, and netting them on a spreadsheet hides both peaks.
No measured funding time. Teams assume a top up is instant because the settlement leg is fast. The slow part is usually the fiat funding, the screening, or a partner cut off nobody wrote down.
Weekend blindness. Payments arrive on Saturday and the funding path does not open until Monday, so the real exposure runs the length of the weekend.
No rehearsal. Run the funding path once on a quiet day, end to end, and time it. That one exercise tells you your real response time instead of your intended one.
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. It settles in regulated stablecoins on public blockchains, in minutes, with on chain auditability, which is what makes a smaller buffer defensible: the funding path and the settlement path run on the same rail and the same clock. It is ISO 20022 native, so a corridor produces pacs.008 customer credit transfers, pacs.009 interbank legs, pacs.002 status reports and pacs.004 returns, tracked end to end by UETR, and a treasury team can tie every movement in the buffer to a message and a transaction hash. Compliance is built in rather than bolted on: KYB and KYC onboarding, KYT, sanctions and PEP screening, FATF Travel Rule data in IVMS101, a compliance workbench and a tamper evident audit trail, so a screening hold shows up as a known event rather than an unexplained gap in the position. Universal Compliance Control, the SpendTheBits submission named a finalist in the Swift Hackathon 2026 Technical Challenge, applies the same discipline to settlement itself: an off chain compliance oracle issues one signed attestation and on chain gates enforce it identically across chains, so no valid attestation means no settlement. SpendTheBits is a Bank of Canada registered payment service provider.
See it on your corridors
Book a working session and we’ll map StableNet’s compliance and settlement to one of your live payment flows.