A payment fee pays for more than moving a message
Merchant acquiring enables businesses to accept card payments and receive settlement. The customer sees a checkout experience; the merchant sees fees, deposited cash and support when a payment is disputed. Behind both is a chain of authorization, processing, settlement and financial responsibility.
For the acquirer and its partners, revenue needs to cover the services supplied and the exposure retained. The OCC’s merchant-processing handbook discusses the distinction between acquiring and issuing cards and the combination of processing costs and contingent loss exposure. Its dated institutional and market descriptions should not be read as current market-share data. [1][2]
The exposure is delayed
Merchant acquiring enables card acceptance and settlement. The OCC’s Merchant Processing handbook treats this as distinct from issuing cards and discusses credit, operational, compliance and other risks. The August 2014 handbook is historical supervisory material, checked through its current official publication page; it is not a new rule. [1][2]
Analysis: a merchant may receive settlement before goods are delivered or disputes are resolved. If customers later obtain refunds or and the merchant cannot reimburse the acquirer, the apparent processing business can become a credit loss. Exposure follows unresolved obligations, not just yesterday’s processed sales.
Three protections that should not be conflated
A rolling reserve retains part of settlement under the contract. A fixed funded reserve is a designated pool. A guarantee or indemnity is a promise whose value depends on enforceability and the counterparty’s ability to pay. Their legal treatment and access conditions differ; a spreadsheet total should not simply add them as equally liquid protection.
The OCC’s payment-processor guidance emphasizes understanding merchants and monitoring volume and information. The practical implication is to connect the reserve calculation to the underlying business, sales pattern and delivery cycle, rather than use the same percentage for every merchant. [3]
Processed volume and gross fees are not profit
Hypothetical monthly program: $10 million of card sales is charged a blended 2.5% merchant fee, producing $250,000 of gross fees. Assume $190,000 is passed through for interchange and network costs, $25,000 supports processing and operations, $20,000 covers sales and customer support, and $10,000 covers net dispute and fraud losses. The remaining contribution is $5,000 before omitted overhead, capital costs and taxes.
If net dispute and fraud losses rise by $15,000 with all else unchanged, the program loses $10,000 on the same sales volume. The assumptions are illustrative, not a market fee quote or a reported processor margin. Actual revenue recognition and contractual allocations vary; a firm reporting net revenue should not be compared directly with another’s gross merchant charges.
Analysis: better authorization and fewer unresolved disputes can help merchants and processors, but not every improvement is captured by one party. Compare completed sales, total merchant cost, support work and losses together. Lowering a visible fee while increasing holdbacks or service friction may shift the economic burden rather than reduce it.
Worked example: a reserve that looks large
Hypothetical: an advance-sale merchant has $2 million of undelivered purchases. Assume a stress produces $600,000 of net valid customer claims after recoveries. A $200,000 funded reserve plus a $500,000 unsecured merchant guarantee is not $700,000 of certain protection. If the merchant fails and the guarantee recovers nothing, the immediate gap is $400,000.
Now assume a 10% rolling holdback on $100,000 of daily sales for 30 days. It accumulates $300,000 before releases, fees and other adjustments—not 10% of all historical exposure indefinitely. Once sales stop, replenishment stops too. The release schedule and remaining claim tail determine whether protection is still adequate.
A reserve decision should explain its assumptions
Analytical control map:
Scroll horizontally to see all columns.
| Input | Question | Consequence |
|---|---|---|
| Unfulfilled sales | What remains undelivered or cancelable? | Measures exposure beyond recent processing |
| Claim timing | When can valid disputes emerge? | Determines the reserve-release tail |
| Recoveries | Which funds are actually controlled? | Separates cash from unsecured promises |
| Concentration | Do merchants share a platform or supplier? | Tests correlated failure |
| Customer friction | Will holdbacks create distress? | Reassesses risk created by the protection itself |
Reserve cash is also merchant working capital
A holdback changes when a merchant receives cash from a completed sale. For a business that must buy inventory or pay installers before the customer receives the goods, that delay can affect its ability to perform. The retained reserve may protect the acquirer while reducing cash available to prevent the underlying non-delivery problem.
Analysis: evaluate reserve size alongside order fulfillment, cash generation and the timing of claims. Clear reporting should tell the merchant what is held, how releases are determined and how disputes affect available settlement, within the actual contract. Verified delivery milestones, slower growth in advance sales or a different settlement schedule may sometimes address the problem more effectively than an unexplained reserve increase. None is a universal substitute for adequate protection.
Reserve tightening can change merchant behavior
Analysis: larger holdbacks can protect an acquirer but also reduce the merchant’s cash to complete orders. A poorly timed reserve increase can therefore worsen the non-delivery risk it was meant to cover. Consider staged restrictions, limits on advance sales and verified delivery milestones alongside collateral.
Delivery delays, refunds, dispute aging, abrupt ticket-size changes and unusual sales acceleration can signal changing exposure. None alone proves fraud, and seasonality or business changes may explain individual observations. A favorable ratio during rapid growth can lag the exposure embedded in unfulfilled sales.
A useful service leaves the exposure and the cash understandable
When processing stops, reconstruct undelivered sales, unresolved claims and the remaining funded protection before treating the relationship as economically finished. The ability to hold or apply funds depends on the contract and applicable law; this article does not prescribe a universal reserve rate or dispute deadline.
Analysis: the acquiring model is stronger when merchants can understand their cash position, customers can resolve valid problems and program contribution remains viable after support and losses. The assessment weakens when rapid growth conceals unfulfilled orders, guarantees replace usable funds in the analysis or tightening reserves makes delivery less likely. The objective is sustainable acceptance and settlement, not simply the largest possible reserve.
Sources
- OCC, Merchant Processing handbook publication page; August 2014Official sourceBack to text: ↑1↑2
- OCC, Merchant Processing handbook; August 2014 edition with March 20, 2025 reputation-risk reference removals, checked September 30, 2026Official source · PDFBack to text: ↑1↑2
- OCC Bulletin 2008-12, payment processors risk management guidanceOfficial sourceBack to text: ↑