Haute Lumière

Commerce · III.09 · MMXXVI · daylight

La Bourse  /  Volume III  /  Nº III.09  /  Workbook — the executive

WORKBOOK — THE CORPORATE EXECUTIVE

Chapter III.09 · Digital Settlement Without Extraction

Applied to a P&L, a board paper, a business unit.

---

THE PREMISE, STATED COMMERCIALLY

You have a cost line that has grown every year, that you have never tendered, that nobody in your organisation owns, and that is priced as a percentage of your revenue. It is payment acceptance. It rose 5.90 per cent across the American market between 2024 and 2025, from $187.20 billion to $198.25 billion, and it will rise again, because it is indexed to your own prices.

Here is the case in one worked example, at a scale a board recognises.

  revenue                                   $ 240.0 million
  card-borne share                                   70 %
  card volume                               $ 168.0 million
  acceptance fee at 1.57 %                  $   2,637,600
  net margin                                        6.0 %
  net profit                                $  14.4 million

Migrate 30 per cent of that volume to a rail costing 0.22 per cent and you save $680,400 — 4.72 per cent of net profit, in year one, with no new customers, no new product and no capital. At full migration: $2,268,000, or 15.75 per cent of net profit.

That is the paper. Everything below is how to write it so that it survives contact with your finance function, your legal function and whoever owns the customer experience.

---

PART ONE — DISCOVERY

Days 1–30: find the money before you propose anything

1.1 — Decompose the fee. Your merchant statement almost certainly shows a blended rate. Get it separated into three components, in writing, from your acquirer:

• Interchange — paid to the issuing bank, set by the network. Typically the large majority of the total.
• Scheme fees — paid to the network itself.
• Acquirer margin — the only part that is genuinely negotiable today.

Most organisations negotiate the third and leave the first two untouched, which is negotiating the smallest number in the stack. The decomposition is the artifact of the first thirty days. Do not proceed without it.

1.2 — Find the ad valorem exposure. Sort your card volume by average transaction value. Your exposure to percentage pricing is concentrated in the large baskets, and this is where the case is strongest: on a representative 1.8 per cent rate, a €40 basket carries €0.72 and a €4,000 basket carries €72.00, while a flat scheme fee of €0.29 carries the same on both — a difference of 248.3 times on the large one. If your business has a long tail of large transactions, you are paying for value rather than for work, and that is where the first migration should go.

1.3 — Find the volume already on a cheap rail. Every organisation has some. Direct debits, bank transfers from trade customers, invoices settled by instant payment. Cost that volume at its actual rate and compare it with the card rate on similar value. You will usually find that the cheap volume is your most valuable customer segment — larger orders, better payment behaviour — and that nobody has ever thanked them for it.

1.4 — Identify who owns the fee. Ask three people who owns the payment acceptance line. Expect three answers or none. The absence of an owner is the reason the line has never been tendered, and appointing one is a decision available to you this week at no cost.

---

PART TWO — THE ARITHMETIC

Days 31–45: build the case in the language of the firm

2.1 — The two ratios, on one page.

  fee as a share of revenue     = acceptance fee  /  revenue
  fee as a share of net profit  = acceptance fee  /  net profit

The first is how the fee is currently discussed, and it sounds trivial. The second is how it should be discussed. In the worked example above the fee is 1.10 per cent of revenue — and 18.32 per cent of net profit. Same fee, and the second framing is the one that gets time on an agenda.

2.2 — The migration model. Three scenarios, no more. Year-one migration at 30 per cent, a mid case, and full migration, with the saving stated in currency and as a share of net profit at each. Put your assumptions on their own line, labelled as assumptions, and name the two that matter most: the achievable migration rate and the all-in cost of the target rail. An unlabelled assumption in a board paper becomes a commitment by the second meeting.

2.3 — The benchmark, and its honest caveat. Your target rate is 0.22 per cent, which is what Pix costs a Brazilian merchant all in. Be precise about what that number contains: it is roughly 975 times the Banco Central's own settlement charge of 0.000226 per cent of value, and that gap is genuine work — fraud, disputes, onboarding, support. The 0.22 per cent is therefore not a floor achievable by wishing. It is the observed all-in cost of a functioning payment business on a rail with no rent in it, and that is exactly why it is the right benchmark and not an aspiration.

2.4 — Price the things you lose. This is the part that makes the paper credible. Migrating away from cards costs you four things, and each has a price:

• The chargeback process. Instant payments are irrevocable. The card rail's 2.2 per cent buys a dispute mechanism you currently use and do not pay for separately. Cost your annual chargeback volume and ask what an equivalent internal process would cost.
• Working-capital float on credit card settlement timing.
• Conversion, if a customer finds the new method harder. Model a 1-point conversion loss against the fee saving and see which wins — often the fee saving does, but you must show the arithmetic rather than assert it.
• Rewards-driven customer preference. Some of your customers are paid to use the expensive rail. That is the Boston Fed finding — the average card household receives $1,133 a year — and they will not give it up for your margin.

2.5 — And what you do not lose. Speed. The instant rails settle in seconds, irrevocably, against a card authorisation that settles in days and a distributed ledger that finalises in 12.8 minutes at best and 60 minutes probabilistically on Bitcoin. If anybody in the room proposes a ledger for domestic settlement, those are the numbers.

---

PART THREE — DESIGN

Days 46–60: the instrument

3.1 — The structure: a merchant acceptance mutual.

The chapter's instrument, because the honest negative applies to you directly: a rail with no margin has no salesforce, so nobody is going to come and build your acceptance for you. You and your peers fund it, at a fraction of what you currently pay a network to do it worse.

• Levy: 0.10 per cent of migrated volume. On a pooled R$1.00 billion that is R$1.00 million a year of funding against R$19.80 million of fees avoided — a coverage ratio of 19.8 times, and over a five-year term R$5.0 million of levy against R$99.0 million avoided.
• Decision inequality, and it goes on the front page: (2.2 % − 0.22 % − 0.10 %) = 1.88 % of migrated volume, retained.
• Sunset in the constituting document on day one. Five years, stepping to zero on penetration targets. A levy without a sunset becomes interchange within a decade and will be defended by people hired using it.
• One member, one vote. You have the volume; the small merchants have the acceptance problem, and the mutual only works if it solves theirs.

3.2 — Balance-sheet treatment. The levy is an operating cost within cost of sales, directly offsetting the merchant discount rate line it replaces. It is margin-accretive from month one and requires no capital approval, which means it does not queue behind your capital programme. The mutual's capitalised estate sits on the mutual's balance sheet; members hold a membership interest at nominal value. Take it to your auditors as a purchasing cooperative, which is what it is, and they will have seen one before.

3.3 — Counterparty sequence. Trade association first — it has the membership list, the legal personality and a collection mechanism. Your acquirer second, because the alternative for them is losing the volume outright. The central bank in year two, with penetration data in hand, so you arrive presenting a mechanism rather than requesting one.

3.4 — What you do unilaterally, this quarter, with no mutual at all. Three moves that need nobody's agreement:

1. Appoint an owner for the acceptance line.
2. Offer your trade and wholesale customers the cheap rail with a discount equal to half the fee you save. You keep half, they keep half, and the migration pays for itself on both sides of the invoice.
3. Re-tender the acquirer margin — the one component that is genuinely negotiable — while the rest of this runs.

3.5 — The three objections your finance function will raise, pre-answered.

"The saving is not real, because the acquirer will reprice." Partly true, and it is why the decomposition in Part One matters: the acquirer margin is the only component they control, and it is the smallest of the three. Interchange and scheme fees are set by the network and will not move for you.

"Our customers will not change behaviour." The evidence says they will if somebody is paid to move them, and will not if nobody is. Brazil reached 54.7 per cent of retail payment transactions in five years under a participation mandate; the United Kingdom reached 11.1 per cent in seventeen with neither a mandate nor a commission. The variable is not consumer willingness. It is whether anybody's job depends on the migration, which is precisely what the unilateral discount and the owner appointment address.

"This is a regulatory bet." It is not. Every number in the case is from an operating rail carrying live volume today, and the instrument — a purchasing cooperative with a levy and a sunset — needs no regulatory change to exist. If regulation moves in your favour it improves the case; nothing in the case depends on it.

3.6 — Sequence the incentive the way India did, and withdraw it the same way. If you fund migration with a customer discount, treat it as a subsidy with a defined end. India paid 0.15 per cent of value on small merchant transactions, spent 8,730 crore over four years, reached 83.4 per cent of ecosystem volume, and only then introduced a 0.4 per cent fee above ₹2,000 with a ₹300 cap on 15 October 2026 — still 5.90 times cheaper than American credit interchange. Spend early, withdraw after the habit is set, and say at the outset that you will. A discount withdrawn on a published schedule is a commercial decision; the same discount withdrawn without warning is a broken promise, and it costs more than it saved.

---

PART FOUR — DESTINY AND DELIGHT

Days 61–90: make it hold

4.1 — Get the ratio into the standing pack. One line, monthly: acceptance cost as a share of net profit, with the migration rate beside it. Anything reviewed monthly persists; anything reviewed by exception does not.

4.2 — Attach it to somebody's objectives. Not necessarily much. An unpaid metric is a hobby, and the migration rate is exactly the kind of number that stalls at 30 per cent and stays there because no one's year depends on it moving.

4.3 — Name the failure modes out loud, in the paper.

• Acceptance stalls because customers were never given a reason. The fix is the shared discount, not a communications campaign.
• The mutual drifts and the levy outlives its purpose. The fix is the sunset, written before anybody is hired.
• Disputes were never funded. The fix is to cost the chargeback equivalent in the original paper, which is why it is in Part Two.
• The subsidy ends before the habit forms. India spent 8,730 crore, reached 83.4 per cent of ecosystem volume, and only then introduced a 0.4 per cent merchant fee above ₹2,000 on 15 October 2026. Had the money run out in year four, the rail would have stalled. Sequence your own incentive the same way: spend early, withdraw only after the habit is set.

4.4 — The delight, and it is commercial. The moment worth having is not the saving. It is the quarter when the acceptance line becomes too small for anyone to model, and your commercial team stops routing decisions around payment cost. The hesitation before saying yes to a small order goes away. Nobody decides to stop hesitating; it simply becomes untrue.

---

THE NINETY DAYS ON ONE PAGE

---

BOARD PAPER TEMPLATE

Title. Payment acceptance: reducing a cost line indexed to our own revenue.

Recommendation. Migrate 30 per cent of card-borne volume to the instant rail in twelve months, funded by a 0.10 per cent acceptance levy through a merchant mutual with a five-year sunset.

The number. $680,400 in year one — 4.72 per cent of net profit — rising to $2,268,000, or 15.75 per cent, at full migration.

What it costs. The levy, at 0.10 per cent of migrated volume; the internal dispute process replacing chargeback; a modelled conversion allowance.

What we lose if we do nothing. The line is indexed to revenue and to inflation. It rose 5.90 per cent across the market in a single year. Doing nothing is a decision to let it compound.

Risks and how they are managed. Acceptance stalls — mitigated by sharing half the saving with the customer. Mutual drift — mitigated by the sunset, in the constituting document, before anyone is hired. Dispute exposure — costed in Appendix B rather than assumed away.

Decision requested. Appointment of an owner for the acceptance line, and approval to sign the mutual's constituting document.

---

APPRECIATIVE QUESTIONS FOR YOUR LEADERSHIP TEAM

1. Which of our customers already pay us on the cheapest rail available — and what have we ever done to thank them?
2. What would we do with 15.75 per cent more net profit that we are not doing now? Answer specifically, before the money is real.
3. Where else in this business do we pay a percentage for something whose cost does not scale with value?
4. If we shared half the saving with our trade customers, what would that change about the relationship beyond the price?
5. What would keep our own levy from becoming, in ten years, exactly the thing we built it to replace — and who in this room would be the one defending it?