A List of Cashflows Is Not a Liquidity Position

A cashflow schedule can be useful and accurate without yet being an operational liquidity position. That position needs an observed balance, certainty, time-bound limits and current movement state.

The difference becomes clear in a small example. One entity has two sterling bank accounts, three cashflows and a routine funding route. The daily numbers look comfortable. The intraday position needs action.

The cashflow schedule says the account is safe

At 08:00, the operating account has an observed balance of £12 million. Three cashflows are due that day:

Read as a daily schedule, the account projects a £6 million closing balance:

£12m - £7m + £6m - £5m = £6m

There is no arithmetic error. The schedule records a useful forecast of the day. It just hasn't answered the operating question yet.

The broader argument appears in Why Treasury Liquidity Looks Like a Digital Twin Problem. Usable liquidity depends on location, time and transfer capability. Here, we can see that argument inside one entity.

The schedule is useful, but a position needs an anchor

A schedule describes movements. An operational position describes usable resources and binding claims at each relevant time.

The operating question is narrower:

What cash remains usable at each required time under the facts currently confirmed?

The £12 million comes from a bank balance observation at 08:00. The observation identifies the account, source and as-of time. It is fresh under the declared window. Treasury has reconciled it to the ledger.

That gives the calculation an authoritative anchor. The next calculation can start from the same observation and apply confirmed changes after 08:00. A later bank observation can replace the anchor and test the earlier projection.

The original schedule still lacks four pieces of operating context. It doesn't say which future cashflows are binding. It doesn't state when account limits apply. It doesn't describe a route from the reserve account. It doesn't track money accepted for transfer.

Not every future pound has the same status

The two payments are committed. They belong in the confirmed curve from the point that they become binding. The expected receipt is different. It belongs beside that curve, not inside it.

The expected receipt shows treasury a possible recovery. It cannot cover a confirmed payment at 08:00. A probability score would not change that control. Ninety-five per cent expected is still not settled cash.

This distinction doesn't make the forecast unimportant. It gives the forecast an honest role. It informs risk and timing without silently turning into available money.

A limit needs a time rule

The operating account has two lower limits:

The hard floor defines whether the account can settle its obligations without unauthorised credit. The operating buffer defines the headroom treasury must preserve under policy.

A buffer isn't fully defined by its amount. It needs an effective window and an evaluation rule. A £3 million buffer tested at every material settlement can produce a different result from the same buffer tested only at close of business.

The live rule in the table governs this calculation. Treasury tests the account after each material settlement. Consuming the buffer triggers action under policy. Crossing the hard floor means the account cannot meet the obligation on the stated terms.

Time exposes the operating-limit breach

Remove the expected receipt from the confirmed curve and retain the two committed payments:

08:00  £12m
10:00   £5m
14:00   £0m

Now apply the live £3 million operating buffer:

At 08:00 and 10:00, the account remains within its operating envelope. By 14:00 it has left that envelope without crossing its hard floor. The positive projected daily closing balance masks this operating-limit breach.

Payroll can still settle. The £5 million payment takes the account to its £0 hard floor, not below it. This is not a hard liquidity breach. A hard breach would put the account below its floor or leave a confirmed obligation unable to settle without unauthorised credit.

Neither outcome is welcome. The hard breach is much more serious. The funding action here preserves required headroom. It is not rescuing a payment that would otherwise fail.

The second account creates an action

The entity has a reserve account with an observed £8 million balance. Its hard floor is £0 and its live operating buffer is £4 million. That leaves £4 million of headroom at 08:00.

The permitted route to the operating account has three relevant terms:

• £3 million remaining capacity
• a 12:30 initiation cut-off
• a 45-minute declared completion time

The route capacity limits the transfer to £3 million. The useful output is:

£3 million can reach the operating account by 13:15, provided treasury acts by 12:30.

That arrival leaves forty-five minutes before payroll.

Treasury can recalculate at the 12:30 decision point. A settled receipt updates the observed position. A receipt that has become binding but remains unsettled can enter the confirmed curve at its revised settlement time. A receipt that remains expected leaves the projected buffer deficit in place. This process can run through scheduled calculations. It does not need continuous event processing.

The source account needs its own check:

£8m observed balance - £3m transfer = £5m projected balance
£5m - £4m operating buffer = £1m remaining headroom

The transfer is feasible at both accounts. It leaves the reserve account £1 million above its buffer and restores the operating account to its required buffer.

The resulting confirmed operating position is:

08:00  £12m observed
10:00   £5m after committed payment
13:15   £8m after confirmed transfer settlement
14:00   £3m after committed payroll

The expected £6 million receipt remains visible beside this curve. It contributes to the confirmed curve only if its state changes to confirmed. It becomes observed cash only after settlement.

A transfer does not move instantaneously

The funding movement follows the same lifecycle used in article one:

Proposed → Accepted → In flight → Settled → Observed state

The proposed movement identifies the amount, route and intended times. Acceptance reserves £3 million of source capacity and the same amount of route capacity. It removes the destination requirement from further allocation, so another calculation cannot fund it twice.

Acceptance does not change the observed bank balance. The reserve account still reports £8 million, but £3 million is no longer usable for another decision.

A confirmed source debit moves the transfer to In flight. The money has left the reserve account, but it has not reached the operating account. Counting it at either end would overstate liquidity.

Destination settlement moves the transfer to Settled. That evidence fulfils the operating-account requirement pending local confirmation. Fresh, reconciled observations then update the Observed state of both accounts.

The lifecycle keeps intention, reservation, debit, receipt and observation as separate facts. That is what stops one transfer from appearing as usable money in two places.

The same facts support other views

The four readings below start from the same balance observations, cashflows, forecast receipt and planned transfer. They do not source a new copy of each balance or cashflow for every horizon. A view can exclude a fact under its rules. Each view applies a different rule set to the shared base layer.

The daily result needs its assumption stated. Treat the receipt as arriving by close of business and test the buffer only at that point:

£12m opening balance
- £7m committed payment
+ £6m expected receipt
- £5m committed payroll
= £6m projected end-of-day balance

£6m projected balance - £3m end-of-day buffer
= £3m projected headroom

That view contains no projected end-of-day buffer breach. The word probably matters. At 08:00, the receipt is still expected and gives no cover in the confirmed intraday curve.

The regulatory row does not claim that these figures prove LCR compliance. They don't contain enough information for that calculation. LCR tests high-quality liquid assets against stressed net cash outflows over 30 calendar days.¹ The temporary breach of an internal operating buffer is not that test.

Basel's LCR calibration does not include intraday liquidity. Separate monitoring tools cover available intraday liquidity, time-specific obligations and payment activity.² The regulatory view needs prescribed reference data and rules on top of the shared facts. It does not need the underlying balances and cashflows sourced again.

The base facts can remain the same. The buckets, classifications and assumptions change with the question.

What turned the list into a position

The first schedule was useful. It became an operational position only after the calculation anchored it to observed cash, separated binding claims from forecasts, applied limits at declared times and represented the funding movement through its lifecycle.

The minimum model is small:

Observed cash
+ binding claims
+ time
+ operating limits
+ transfer capability
+ movement state
= operational liquidity position

The same base records can support intraday control, daily planning and regulatory derivations. That does not mean one calculation or one output serves every purpose. It means each view starts from the same economic events and states its own rules.

The worked position covers one entity, one currency and two accounts. More accounts, currencies, entities and jurisdictions add route, ownership and aggregation constraints. Failed, partial and reversed transfers add lifecycle branches. None of those extensions changes the starting point.

A cashflow list tells treasury what is expected to move. An operational liquidity position tells treasury what remains usable, which claims bind it and what action can still complete in time.

References