The Card Acquiring Margin Compression Trap

A card acquirer's economics look simple from the outside: collect MDR from the merchant, pass through interchange, keep the margin. The margin is thin for well-priced enterprise accounts — no acquirer publishes what it actually nets, and the figure varies by merchant segment and contract — but the volume is large. At sufficient scale, thin margin on enormous GMV produces a real business.

What's changed over the past decade is that the forces determining each component of that margin are all moving in the wrong direction simultaneously, and they're not moving back. Understanding why — and who survives it — is essential context for any merchant evaluating long-term PSP relationships or any operator building payment infrastructure.

The Acquiring Margin, Decomposed

Start with the actual economics of a card acceptance transaction. Take a UK consumer Visa credit card transaction for £100. The one line item here that is regulator-confirmed and publishable: under the EU/UK Interchange Fee Regulation, consumer credit interchange is capped at 0.3%, so the issuer receives £0.30.

What the merchant pays in MDR, what Visa keeps in scheme fee, and what the acquirer nets after processing costs are not published anywhere — they're commercially negotiated per merchant and per acquirer, and vary with volume, risk profile, and contract terms. The regulated £0.30 interchange leg is a fixed, known cost; the scheme fee and the acquirer's own margin are the two variables an acquirer actually controls, and the two a merchant should be pushing to see broken out as separate line items rather than folded into a blended rate (see interchange-plus vs blended pricing). Treat any specific scheme-fee or acquirer-margin figure you're quoted as a negotiated number, not a market rate — nobody publishes one.

That's the regulated-interchange market. In the US, there is no equivalent cap on consumer credit card interchange, and rates vary by card tier and merchant category rather than sitting at one published figure — but the interchange leg is set independently by each card-issuing bank rather than fixed by regulation, leaving less structurally-guaranteed room for acquirer margin at the same MDR.

The acquiring margin sits in the space between what the merchant pays and what the networks extract. Both forces are squeezing that space.

The Interchange Cap Ratchet

The EU's Interchange Fee Regulation (IFR), effective 2015, capped consumer debit card interchange at 0.2% and consumer credit at 0.3%. These caps apply only to consumer cards; commercial and corporate cards are explicitly excluded and uncapped. This was the first major regulated interchange cap in a large market. The UK's Payment Systems Regulator was actually established earlier and for a broader remit than IFR enforcement — created under the Financial Services (Banking Reform) Act 2013 and operational from April 2015, more than a year before the Brexit referendum — and was later designated as the UK's enforcer of the IFR caps rather than founded for that purpose. The UK retained the IFR caps post-Brexit, with the PSR as the body monitoring compliance.

Australia's current consumer and commercial credit interchange regime caps individual transactions at 0.80%, with a separate 0.50% weighted-average benchmark on the total credit interchange each card network charges — not a single 0.50% cap. The RBA's 31 March 2026 Conclusions Paper abolishes the benchmark and gives consumer credit a flat 0.30% cap from 1 October 2026. Commercial credit keeps its 0.80% cap, and the RBA's own figures show it already averaging 0.78%, so commercial interchange barely moves on average — though networks may raise strategic rates that sit below the cap — while consumer credit gets markedly cheaper — the acquirer margin question sits increasingly on the commercial side of the mix. A separate measure affecting foreign-issued card interchange is planned for 1 April 2027 — a different date, not to be conflated with the consumer credit change.

The US Durbin Amendment caps debit interchange at $0.21 plus 0.05% of transaction value, plus a further $0.01 fraud-prevention adjustment for issuers that meet the Fed's fraud-prevention standards, for regulated issuers (banks above $10B in assets). Exempt issuers (community banks, credit unions) can charge higher rates. Durbin has been in place since 2011; there are periodic discussions about extending regulation to credit cards that have not yet succeeded but represent a policy direction.

The pattern: once a jurisdiction regulates interchange, subsequent reviews trend toward lower caps, not higher. No major regulated market has seen interchange caps rise after initial regulation. The direction of travel is one-way.

For acquirers operating in regulated markets, this means the gross revenue pool — total MDR across all transactions — compresses over time. Each cap reduction translates directly to either MDR pressure on merchants (who benefit and accelerate pressure for further caps) or margin erosion for acquirers (who can't reduce costs fast enough to offset the revenue reduction).

Rising Network Fees

Visa and Mastercard don't receive interchange — they receive scheme fees, which they set independently. As interchange fell under regulation, scheme fees rose. In the UK specifically, the Payment Systems Regulator's market review (MR22/1.10) found that Visa and Mastercard scheme core scheme and processing fees to acquirers grew by more than 30% in real terms over 2017/18–2021/22 — a UK-only finding, not a global one — with increases concentrated in card-not-present transaction types where the networks have the most pricing power.

This is structurally rational from the networks' perspective. They don't receive interchange; they're not harmed when it's capped. They can raise their own fees to capture value that regulation removed from the interchange line. Regulators have been slower to address scheme fees than interchange — the EU is investigating Visa and Mastercard scheme fees, but enforcement action is years behind the interchange work.

For acquirers, rising scheme fees mean that even at stable MDR, the amount left after passing through the regulated interchange floor and the rising scheme fees shrinks. Whatever an acquirer's scheme-fee line item costs today, the PSR's UK finding above says it was materially lower as recently as 2017/18.

Real-Time Rails Taking Volume at Zero

The most structurally significant long-term pressure is different from both of the above: account-to-account real-time payment rails operating at zero or near-zero MDR.

Pix in Brazil has grown to a projected ~8 billion monthly transactions by year-end 2025, according to an EBANX release (14 November 2025) — though that figure is EBANX's own projection, not a confirmed year-end count. What Pix has actually displaced is more nuanced than a simple card-volume-share story: BIS research (BIS Papers No. 152) finds Pix has mainly displaced cash, with merchant card-acceptance volume continuing to rise alongside it rather than being cannibalised. Pix is genuinely zero-cost for P2P transfers, but merchant-facing Pix — Pix Cobrança and QR checkout — is not free: PSPs commonly charge merchants 0.99–1.89% to accept it, so the true merchant-cost comparison against card MDR is narrower than a headline "zero-cost" framing suggests. In India, UPI processed 21.63 billion transactions in December 2025 alone (NPCI) at zero merchant discount rate — a policy outcome of a 2019 amendment to the Income-tax Act (Section 269SU) rather than an RBI rule, and one that remains actively contested: as of August 2026 there is live policy debate in India over reinstating MDR above ₹2,000. In Australia, NPP/PayID processes billions annually at near-zero merchant cost. Faster Payments in the UK handles bill payments, A2A transfers, and an expanding set of merchant payments.

This isn't a short-term phenomenon. These rails are growing faster than card volume in the markets where they operate. The total addressable market for card acquiring is growing more slowly than overall payment volume — the rails are taking the incremental.

For acquirers, this means that the volume growth that historically justified scale investment and absorbed margin compression is no longer available in the same form. A Brazilian acquirer that was planning for 15% volume growth is now planning for 7% card volume growth with the remainder captured by Pix.

How Stripe and Adyen Escaped

Stripe and Adyen both recognized this structural problem early and pivoted toward software revenue before the margin compression fully materialized.

Stripe is privately held and discloses no revenue or margin breakdown, so any specific split between payment acceptance and software products is not verifiable and shouldn't be treated as fact. What is observable from Stripe's own product catalogue: alongside core payment acceptance, it has built out Billing, Revenue Recognition, Tax, Sigma (analytics), Radar (fraud), Connect (marketplace payments), Issuing (card issuing), and Treasury (banking-as-a-service) — software products that plausibly carry higher gross margins than payment processing and create switching costs that pure-play acquiring doesn't generate. The strategic logic is visible in the product lineup even though the financial mix behind it is not public.

Adyen follows the same model at the enterprise end: unified commerce platform premium (one integration for online, in-app, and POS), data-as-a-service for enterprise analytics, and embedded financial products for platform and marketplace clients. Adyen's take rate is lower than Stripe's because it serves larger merchants who negotiate harder, but its per-account revenue is higher because the software attachment rate is significant.

The strategic move: use payment processing acceptance as a loss-leader that creates the customer relationship, then sell increasingly sticky software products into that relationship. The acquiring margin compression affects the loss-leader; it doesn't affect the software products.

The Legacy Acquirer Problem

FIS, Fiserv, WorldPay (majority stake sold by FIS to PE firm GTCR in February 2024, with FIS retaining a minority position), Global Payments, TSYS — the enterprise acquiring incumbents — built their businesses in the era when acquiring margins were 0.5%+ and volume grew predictably. Their cost structures include: large direct sales forces, heavy implementation services, complex billing systems, and legacy mainframe processing infrastructure. These are high-fixed-cost businesses.

The margin compression doesn't affect their costs proportionally. A 10% reduction in acquiring margin doesn't reduce sales force costs or mainframe maintenance costs. It reduces net income.

Their response has been acquisition: FIS acquired SunGard, IFS, Certegy, and WorldPay in successive deals. Fiserv acquired First Data. WorldPay acquired Vantiv, which had previously acquired many smaller acquirers. The thesis: scale reduces per-unit cost, and acquired companies bring volume. The problem is that the acquisitions were priced at premium multiples assuming continued acquiring margin, and the integration costs and debt service have complicated the balance sheet exactly as margins are compressing.

FIS acquired WorldPay for approximately $43 billion in 2019, then took impairment charges of $17.6 billion in 2022 and $6.8 billion in 2023 against that goodwill. In February 2024, FIS sold a 55% majority stake in WorldPay to private equity firm GTCR at an $18.5 billion valuation (roughly $11.7 billion in net proceeds to FIS), retaining the remaining 45% itself. Even before accounting for the earlier impairments, the $18.5 billion valuation of the whole business was less than half what FIS paid five years earlier. This is the most explicit signal of the structural problem: the largest acquiring asset in the world was valued, in a fresh arm's-length transaction, at a fraction of its acquisition price.

What This Means for Merchants

The merchant implications are direct:

Negotiate IC+ now. Blended pricing shields merchants from seeing how acquiring margin is changing. Interchange-plus pricing makes the margin visible and negotiable. No acquirer publishes a volume threshold for when IC+ becomes available or what margin to expect at it — those terms are negotiated case by case — but the practical move is the same regardless of your volume: ask for interchange, scheme fee, and PSP margin broken out as separate line items, and treat the margin line as the one that's actually negotiable.

Watch for scheme fee pass-throughs. As acquirers face margin pressure, some are renegotiating merchant contracts to pass through scheme fee increases. Review your contract for language on scheme fee changes — specifically whether the acquirer can unilaterally increase fees to reflect scheme fee changes, and whether there's a cap or notice requirement.

Maintain optionality. The legacy acquirer consolidation creates integration risk. When FIS sold WorldPay, merchants experienced months of pricing and support uncertainty. Merchants with a single acquirer relationship who rely on that acquirer's continuity are more exposed to corporate actions affecting their PSP. Two acquirer relationships at meaningful volume significantly reduces this risk.

Monitor real-time rail MDR actively. In markets where real-time rails have near-zero MDR — Australia, Mexico (SPEI), UK (Faster Payments for bill pay) — specific transaction types are already cheaper to route via the rail than via card. The analysis of which transactions to route where is worth doing annually, not just at contract renewal.

The acquiring business is not going to zero — cards remain dominant for consumer commerce in most markets and will for a decade. But the pure-play acquiring margin trajectory is structurally negative, and the operators who understand that are positioning accordingly.