Interchange Regulation by Market: What the Caps Actually Cover

A regulated interchange cap reads like a promise: card acceptance in this market costs no more than X%. Operators build pricing models on that promise, then find their effective rate running well above the cap they thought applied — not because the regulator failed, but because the cap was never the whole story. It caps interchange — one of three layers in the MDR — and in every regime covered here, it excludes something that matters: commercial cards, cross-border transactions, or an entire card type.

This is a market-by-market reference to seven regimes — EU/EEA, UK, US, Australia, Brazil, India, and South Africa — chosen because each is independently sourced to a regulator, court filing, or dated legal commentary, not because the list is exhaustive. What matters operationally is not memorising a rate table that will be stale within a year, but understanding the mechanism each regulator uses and where its edges are, so you can predict what won't show up in the cap before your acquirer's invoice tells you.

The Three Layers, and Which One Gets Regulated

Card scheme fees demystified covers the full anatomy: MDR decomposes into interchange (paid to the issuer), scheme fees (paid to Visa or Mastercard), and acquirer margin (kept by your PSP). Every regime in this article regulates interchange. None of them regulates scheme fees, which the networks set unilaterally and which have grown 20-30% since 2019. None of them regulates acquirer margin, which is a commercial negotiation covered in interchange-plus vs. blended pricing. A merchant in the most tightly capped market in this list can still watch their effective rate climb if scheme fees rise or their card mix drifts — the cap was never protecting against either.

Within interchange regulation itself, five distinctions determine whether a given transaction is actually covered by a cap an operator assumes applies:

• Regulated cap vs. MDR. A cap on interchange is not a cap on what the merchant pays. Acquirer margin and scheme fees stack on top, uncapped, in every market below.
• Debit vs. credit. Every regime here treats them differently — sometimes one is capped and the other isn't at all (Brazil, effectively the US).
• Domestic vs. cross-border. The UK is the sharpest example: the same cap, applied to a narrower set of transactions than it was before 2021, with cross-border fees now running five to six times higher.
• Consumer vs. commercial/corporate cards. Commercial cards are exempt from the EU/UK cap outright, and in Australia's 2026 reform keep their 0.80% ceiling number while losing the weighted-average benchmark that held realised commercial interchange below that ceiling.
• Domestic-issued vs. foreign-issued cards. Australia is introducing its first-ever cap on foreign-issued cards in 2027; before that date, those transactions are entirely unregulated even though domestic debit and credit are tightly capped.

EU/EEA: The Reference Model, With a Gap for Tourists

Regulation (EU) 2015/751 caps consumer debit interchange at 0.2% and consumer credit at 0.3% of transaction value, effective from 9 December 2015. Three-party schemes (Amex, Diners) had a three-year exemption running to 9 December 2018. Commercial cards — issued to a business, public-sector entity, or self-employed person and limited to business expenses — are excluded from the Regulation entirely, so a corporate card transaction inside the EU carries none of the IFR's protection.

The less-discussed gap is inter-regional: the IFR only governs transactions within the EEA. A card issued outside the EEA used at an EU merchant — a tourist's card, most commonly — was never subject to the 0.2%/0.3% cap at all. The European Commission addressed this separately, not through the IFR but through a 2019 antitrust settlement: Visa and Mastercard committed to cap inter-regional interchange at 0.2%/0.3% for card-present transactions and 1.15%/1.5% for card-not-present. Those commitments were extended on 8 July 2024 to run until November 2029. The EU's cross-border protection, in other words, is a voluntary antitrust commitment with an expiry date — not statute — running in parallel to the IFR's binding domestic cap.

UK: The Cap Held, the Border Moved

The UK onshored the IFR after Brexit at the same 0.2%/0.3% consumer caps, but narrowed its scope: the cap now applies only where merchant, acquirer, and card issuer are all within the UK. A transaction that used to qualify as domestic-equivalent under the EU-wide IFR — a UK merchant serving an EEA cardholder — became cross-border the moment the UK left, and cross-border transactions are not covered by either the UK's onshored cap or the EU's.

Visa and Mastercard responded by raising UK-EEA cross-border interchange on card-not-present transactions from 0.2% to 1.15% (debit) and 0.3% to 1.5% (credit) — a roughly six-fold and five-fold increase respectively. The PSR's market review found no competitive constraint justified the increase and estimated the cost to UK businesses at £150-200 million a year. On 15 January 2026, the High Court confirmed the PSR has the legal power to cap these fees, rejecting a joint challenge from Visa, Mastercard, and Revolut — but as of that ruling, no cap had actually been set. The UK is, right now, a market with a firm domestic interchange cap and an uncapped cross-border fee running at multiples of it, with the regulator holding confirmed authority to fix that and not yet having used it.

United States: A Debit-Only Cap With Its Legal Footing in Question

Regulation II caps debit interchange at $0.21 plus 5 basis points of transaction value, plus an optional 1-cent fraud-prevention adjustment for issuers that meet fraud-prevention standards — but only for issuers with $10 billion or more in consolidated assets. Below that threshold, debit interchange is unregulated and routinely runs well above the capped rate. Credit card interchange is not regulated under Regulation II at all, and never has been. The $10 billion threshold governs the fee cap specifically; Reg II's separate requirement that issuers enable at least two unaffiliated debit networks applies to every issuer regardless of size — a network-choice mandate, not a cost cap, covered in full in least-cost routing.

What operators modeling US debit economics need to know now: Regulation II's legal foundation is under live challenge. On 6 August 2025, a North Dakota district court vacated the rule entirely in Corner Post v. Federal Reserve, ruling the Fed had included cost categories — fixed processing costs, fraud losses, network fees — beyond what the Durbin Amendment permits. The court stayed its own vacatur specifically to avoid an unregulated interim market, so the existing cap remains in effect while the case proceeds. Briefing at the Eighth Circuit Court of Appeals concluded in March 2026 with no ruling yet as of this writing. The cap you're pricing against today is the same cap that a federal court has already found unlawful, held in place only by a stay.

Australia: Reform Finalised, Effective in Weeks

The RBA's Review of Merchant Card Payment Costs and Surcharging Conclusions Paper, published 31 March 2026, finalises a package taking effect 1 October 2026 — and the way it treats commercial credit is the single most consequential detail in this article, because reading it as a no-op will misprice an operator's Australian commercial-card volume.

Today's regime is two-part for both consumer and commercial credit: an 0.80% per-transaction ceiling, plus a separate 0.50% weighted-average benchmark that constrains an issuer's realised interchange, averaged across its whole card portfolio, to below that ceiling. The Conclusions Paper abolishes the weighted-average benchmark entirely — for both card types, not just consumer. Consumer credit gets a new flat 0.30% per-transaction ceiling with no benchmark behind it at all. Commercial credit keeps the 0.80% ceiling number unchanged, but loses the averaging discipline that previously held realised commercial interchange below it. The headline figure is the same; the constraint that made that figure meaningful is gone. An operator with material commercial-card volume into Australia should stop modelling at today's blended commercial rate and start modelling closer to 0.80% from 1 October 2026 — the RBA's own Conclusions Paper describes this, card type by card type, as "retaining the interchange cap ... at 0.8 per cent ... and removing the weighted-average benchmark."

Domestic debit and prepaid interchange tightens on the same date, and the current headline understates it: today's cap is itself compound — 10 cents and 0.20% of transaction value per transaction, not just 10 cents flat — falling to 8 cents and 0.16% of transaction value from 1 October 2026. Unlike credit, debit and prepaid keep their own weighted-average benchmark (also at 8 cents), so the averaging discipline stays in place for that card type even as the ceiling tightens. The same package lifts Australia's long-standing prohibition on card-network no-surcharge rules from the same date, a separate but connected change.

The genuinely new element is a cap on foreign-issued cards — cards issued outside Australia, used at Australian merchants — which have never been regulated before. From 1 April 2027, a unified 1.00% cap applies across card type and channel. Until that date, every foreign-issued card transaction acquired in Australia sits entirely outside RBA interchange regulation, regardless of how tightly domestic cards are capped.

Brazil: Debit and Prepaid Capped, Credit Left Alone

Resolução BCB nº 246/2022, effective 1 April 2023, caps the tarifa de intercâmbio at 0.5% per transaction for debit cards and 0.7% for prepaid cards — figures corroborated by multiple independent secondary write-ups of the resolution but not independently retrieved from bcb.gov.br itself this session (its site returns an empty single-page-app shell to fetch tools, and the PDF path 404s). Credit card interchange is not covered by this cap at all — Brazil regulates two of its three major card categories and leaves the third to the market. The 2022 resolution also closed a carve-out in its predecessor, and that predecessor was itself more structurally interesting than a flat cap: Circular BCB nº 3.887/2018 paired a 0.5% weighted-average interchange benchmark (calculated quarterly across transaction value) with a separate 0.8% per-transaction ceiling — the same benchmark-versus-ceiling structure Australia uses for its own commercial credit cap — and excluded corporate cards and card-not-present transactions from both limits. The current resolution applies to all debit transactions without exception, corporate or otherwise, but replaced that two-part mechanism with a single flat 0.5% ceiling rather than preserving a benchmark-plus-ceiling design. Brazil is the one regime in this list that moved in the opposite direction from the EU, UK, and Australia on the commercial-card question — closing a commercial-card exemption rather than preserving one. For the licensing structure sitting underneath Brazilian acquiring, see PSP licensing and local acquiring in Brazil.

India: No Interchange Cap — a Zero-MDR Mandate Instead

India doesn't regulate interchange as a distinct line item at all. The RBI instead caps the merchant discount rate directly for standard debit card transactions, tiered by merchant turnover: up to 0.40% (physical/online) or 0.30% (QR), capped at ₹200 per transaction, for merchants under ₹20 lakh annual turnover; up to 0.90% or 0.80%, capped at ₹1,000, for merchants above that threshold. Banks must ensure merchants don't pass MDR through to customers as a surcharge.

RuPay debit cards and UPI sit entirely outside even that structure: they have run at a statutory zero MDR since January 2020, under Section 10A of the Payment and Settlement Systems Act, 2007. That statute is no longer settled ground — and it is no longer merely proposed to change, either. The Taxation and Other Laws (Amendment) Bill, 2026 cleared the Lok Sabha on 6 August 2026 and the Rajya Sabha on 10 August 2026, then received presidential assent on 17 August 2026, becoming enacted law. The amendment replaces Section 10A's blanket zero-charge mandate with a provision letting the Central Government specify by notification which payment modes may attract a fee. It remains an enabling power, not a fee: no MDR has actually been notified as of enactment, Finance Minister Nirmala Sitharaman stated on the record that no MDR framework "has been finalized as yet," and RBI Governor Sanjay Malhotra had already called reintroduction talks "very premature" at the RBI's 5 August 2026 press conference, before the bill had even finished clearing Parliament. But the legal wall that made UPI's zero-MDR model untouchable by ordinary policy — previously requiring a fresh Act of Parliament to remove — is now gone; a government notification is all a future MDR would take. India's UPI at scale covers the underlying sustainability debate this amendment responds to.

South Africa: A Seventh Regulated Regime, With No Domestic Scheme to Route Around

SARB — the South African Reserve Bank — publishes a regulated interchange determination directly, the same regulator-sets-the-rate model as the RBA or the Fed, but with more tiers than either: the current determination (last updated 5 February 2026) sets debit card interchange at 0.44% of transaction value for card-present, EMV-compliant purchases and 0.58% for card-not-present, 3D-Secure-compliant purchases, rising to 0.98% where only the issuer (not the acquirer) is 3D-Secure compliant. Credit card interchange runs 1.48% and 1.68% for the same tiers, up to 2.45% on the issuer-only-compliant case. ATM cash withdrawal interchange is separately capped at R5.31 plus 0.64% of the withdrawal amount. None of these figures collapses to a single flat rate — an operator quoting "South African interchange" as one number is already flattening a determination that is deliberately tiered by compliance posture and channel. Unlike France, Brazil, or Germany, South Africa has no domestic card scheme: Visa and Mastercard dominate directly, so a foreign acquirer doesn't need separate domestic-scheme routing economics the way it would in those markets — but it also means South African merchants have no domestic-scheme lever to negotiate against these SARB-set rates. See South Africa for the underlying determination and SARB's ongoing Interchange Determination Project Phase V review of the methodology.

Japan is the counter-example worth knowing precisely because it looks similar from a distance but isn't: the JFTC and METI intervened on card interchange in 2022 too, but chose disclosure over a ceiling — Visa, Mastercard, and UnionPay now publish their standard interchange rates in Japan, and nothing caps what those published rates can be.

Seven Regimes, Seven Different Edges

Every figure above is dated to the source cited; several are mid-transition (UK, Australia, India, and South Africa's payment-system governance all have live regulatory or legislative action as of this writing) and should be re-verified against the linked source before being used in a contract negotiation. Japan is deliberately excluded from this table — it discloses standard interchange rates rather than capping them, which is a different regulatory choice, not a missing data point.

What This Means for Operators

Stop pricing to the headline cap — Australia's commercial-card number is the sharpest example of why. If your card mix includes commercial or corporate cards, model them separately. The EU/UK exempt them outright. Australia's 2026 reform leaves the commercial credit ceiling at 0.80% — nearly triple the new 0.30% consumer rate — but the more important change is invisible in that headline: the 0.50% weighted-average benchmark that previously held realised commercial interchange below 0.80% is being removed on the same date the consumer cap falls. An operator with material commercial-card volume into Australia should stop modelling at today's blended commercial rate and start modelling closer to 0.80% from 1 October 2026 — the ceiling didn't move, but the floor under it just fell away. A blended assumption that "this market caps interchange at X%" will understate cost the moment a commercial card shows up in the mix, a problem interchange-plus pricing exists specifically to surface.

Cross-border volume is where capped markets stop being capped. The UK is the clearest case — a domestic cap holding at 0.2%/0.3% while the cross-border rate on the same card networks runs five to six times higher, with no fix yet in force. Any operator with material cross-border card volume into a "regulated" market should check whether the regulation actually reaches that specific flow before budgeting against the domestic figure.

Treat every cap in this article as dated, not fixed. Four of seven markets have live change in progress: the UK's cross-border cap is authorised but unset, the US cap is vacated-but-stayed pending an appellate ruling, Australia's reform takes effect 1 October 2026 with a further foreign-card cap in 2027, and India's zero-MDR statute was just amended — and has now actually been enacted, not merely proposed — to allow a change it previously prohibited by law. A rate correct today can be wrong by the time a contract renews. Card acquiring margin compression covers the structural direction these caps are moving in and what it means for acquirer economics on the other side of the same trend.

The regulatory mechanism is never the whole cost. It's the floor one layer sits on — and in every market here, at least one card type, one transaction direction, or one entire fee category sits above it, uncapped.