# The Executive Case

FNA Sampo coordinates when the banks in a cohort submit their large-value payments,
so that each bank funds a lower intraday peak without changing what it pays,
whom it pays, or who decides when it submits.

> This document describes Sampo's intended target operating model for banks. The supporting results currently come from a synthetic validation
> environment; they demonstrate the mechanism and inform pilot design, but do not
> represent production performance or a forecast of any bank's outcome.

> **The proposition:** measure whether Sampo's payment schedule can reduce your modeled
> in-scope peak net debit, without changing payment amounts, counterparties, rail
> choice, settlement finality, or your authority to submit.

## The problem

Fedwire settles about **$4.6 trillion a day** and runs no coordinating mechanism.
More than half that value settles in the final third of the day, because each bank
waits for incoming funds before submitting its own payments. The peak a bank funds is
therefore larger than its payment obligations require: part of it exists only because
submission times are uncoordinated. Liquidity here is not scarce; its use is uncoordinated.

No bank can capture this alone. The saving is a network property (incoming payments
that, submitted in a better order, fund your outgoing ones instead of your own balances
and collateral), so no single bank acting alone can measure or realize it.

## What Sampo does

Sampo issues **confidential, non-binding advice** for a bank's own
eligible payments, those above an agreed value threshold. Each bank sends payment
intents (the amount, counterparty bank, priority, earliest time and own time of a
payment it will make; not payment instructions) from its own payment system or
intraday-liquidity system. In Sampo's design, independent compute parties compute the
payment schedule together, and each sees only shares of the amounts. In that design, no single party recovers a payment amount or the liquidity
position derived from it. Each bank
receives advice for its own payments only, and Fedwire settles
every payment gross and in full, unchanged.

## What the bank retains

- **Submission authority.** Sampo never submits or holds a payment. Every
  submission remains the bank's own act through its existing systems.
- **Customer data.** Sampo receives each payment's amount, counterparty bank, priority
  and time window. Customer data and personal information stay with the bank.
  Sampo advises on timing only, and never later than the payment's own time.
- **The settlement relationship.** Sampo takes no part in settlement, holds no
  funds, and cannot block or delay Fedwire settlement. Finality is central-bank money
  on the existing rail, exactly as today.
- **The ability to bypass.** A bank need not follow any advice; it can self-fund under
  its own controls, or stop sending payment intents, at any time, without operational
  dependency on Sampo.

## What the evidence shows

On a Fedwire-referenced synthetic model of a five-bank core, Sampo's payment schedule
reduces the cohort monthly peak net debit (modeled, in scope) by
**49.3%, about $55.4bn** on the modeled book. It is synthetic: it
demonstrates the mechanism and sizes the pilot, and it is not a forecast of any bank's
outcome. Sampo's no-worse-off guard aims to reduce the risk that a bank is worse off: that
its realised peak net debit with Sampo ends above its peak without Sampo. It is not a
guarantee. The production precedent for the mechanism class is CHAPS, where the
liquidity-saving mechanism cut settlement banks' intraday liquidity requirement by
roughly **20%, about £4bn a day**, with no loss of throughput. The gridlock-resolution
algorithm underlying that class of mechanism was formalized by Bech and Soramäki (2001)
in [*Gridlock Resolution in Interbank Payment Systems*](https://econpapers.repec.org/paper/zbwbofrdp/rdp2001_5f009.htm).

A lower modeled peak does not automatically release a regulatory buffer, HQLA or
capital, and we make no such claim. What a reduction is worth depends on your own
recognition methodology and your cost of liquidity, an annual rate, both of which
are yours rather than ours. [Data and methodology](../data-and-methodology/) sets out how the modeled figure is
measured and what the model leaves out.

### How your bank validates this

An evidence pilot measures the opportunity on your own flows:

1. FNA replays your agreed in-scope historical or shadow flows.
2. FNA compares your peak without Sampo with your peak with Sampo, under an
   agreed measurement convention.
3. Test adherence and exception scenarios, including the days coordination helps
   least.
4. Apply your own liquidity, internal liquidity stress testing (ILST),
   funds-transfer-pricing and regulatory methodology to any reduction.
5. Decide whether the measured opportunity justifies advisory production.

You get your own bank's figure across the sample, not a network average and not our
number applied to you. If it is not material for you, you learn that early and
cheaply.

## The next step

Agree the scope and success criteria for an evidence pilot: which payments are in
scope, the measurement convention, the adherence assumption, and the threshold at
which a measured result justifies going further. Moving to live operation on real data is a
separate step, with its own approval.