2026 Visa Fee Hike: Break-Even Up, Small Merchants Route to PayTo

Before the Durbin Amendment, a debit swipe cost merchants $0.44 on average; no cap existed. Now Visa's 2026 fee restructure is resetting the calculus for small merchants weighing PayTo. The break-even point has moved up, and for a small transaction, PayTo still loses to Visa on price.

The Visa/Mastercard $38 billion settlement was framed as relief after two decades of interchange litigation, but the fee outlook for 2026 remains opaque. Merchants paid interchange equal to 1.49% of total sales in 2016, and a Federal Reserve study found 22% of merchants raised prices after Durbin. Routing decisions now hinge on exact ticket sizes: a small-ticket PayTo route is slightly more expensive than Visa, while a larger average ticket flips the gap in PayTo's favor.

For a store processing thousands of orders a month, the difference compounds. But the baseline remains clarity: merchants are not asking for zero interchange; they want predictability. With Illinois exempting interchange on sales tax and gratuities, and capping collection compensation at $1,000 a month, states are entering a regulatory patchwork in 2026. Account-to-account routing such as PayTo becomes a viable alternative only after merchants map their own average ticket against the new Visa fee tables.

Why Visa's 2026 Fee Stack Puts a 66¢ Flat Line Next to

Visa's 2026 Rate & Interchange Guide for Australia, effective 24 April 2026, puts a representative small-merchant all-in cost at 1.40% + 20¢ per domestic approval — but the structure matters more than the total. According to the Guide, that representative cost splits into consumer-credit interchange of 1.21% + 8¢, a Visa network fee of 0.19% + 6¢, and a new fixed Authorization Integrity Fee of 6¢. The novelty is not the size of the percentage; it is the new fixed 6¢ line sitting next to the ad valorem charges. That line is what makes a flat 66¢ competitor viable above a specific ticket size.

Because the 6¢ Authorization Integrity Fee is fixed per approval, it does not scale with the order. Percentage fees shrink in relative weight as the ticket grows; fixed fees do not. Every Visa approval now carries 20¢ of fixed cost (8¢ interchange + 6¢ network fee + 6¢ Authorization Integrity Fee), so the 2026 break-even is higher than previous calculations that only modeled the percentage components. This is why the route switch is value-dependent: it is not a uniform "Visa got more expensive" story, but a shift in the cost curve's shape.

The alternative rail is genuinely flat. PayTo is AP+'s account-to-account request-to-pay product built on the New Payments Platform; settlement runs through the RBA's Fast Settlement Service, and the transaction never creates a Visa PAN, card token, or CID. On the cost side, according to Cuscal's published PayTo API pricing, the Core tier charges 66¢ per successful payment for merchants processing under 10,000 transactions per month. There is no interchange or network component, so the merchant's PayTo cost is the same regardless of the order's size.

The checkout mechanics explain why the fallback rule is structural, not optional. At checkout, the PayTo rail sends a Payment Initiation Request to the customer's banking app; the customer approves the exact amount in the app, funds move to the merchant account, and checkout resumes as a confirmed paid order. No card credentials change hands, and the merchant learns of success only after the customer approves in the banking app.

That approval step is also the edge case. A PayTo-first checkout must retain Visa as the fallback: if the customer's bank does not support PayTo, or the user abandons the app approval flow, the browser must return to the card field rather than trapping the sale. The cost comparison makes the boundary explicit. At a $10 order, Visa charges 1.40% × $10 + 20¢ = 34¢, which is cheaper than PayTo's 66¢; at a low-ticket order, Visa remains marginally cheaper. The crossover lands at the break-even threshold shown in this guide's decision table, so orders above that threshold should route through PayTo while Visa remains the default for smaller tickets and all cross-border card payments. The myth that the 2026 fee hike makes PayTo cheaper for every transaction fails precisely because of the fixed-fee increase: it pushes the break-even up, leaving orders below the threshold cheaper on Visa.

The Numbers

The RBA's 2023 Merchant Payments Costs Survey puts the average card-payment cost at 0.83% of transaction value and the average NPP account-to-account payment at 0.11%. That is the structural cost gap between card rails and bank rails, and it is the raw material for the routing threshold in this guide. The NPP figure is not a teaser rate; it is the RBA's own bookkeeping for what a New Payments Platform transfer actually costs once scheme and interchange layers are stripped away. Card rails carry interchange, scheme fees, and fraud pools; the NPP clears directly against a bank account with no interchange layer.

The flat-fee side is the binding constraint. According to Cuscal's published PayTo API price sheet, dated 1 January 2026, the Core tier lists 66¢ per successful payment, with no monthly minimum for volume under 10,000. That 66¢ is invariant to ticket size, which is what makes it a clean benchmark. According to Visa's 2026 Australian Merchant Fee Rates table, the fixed-fee share of the all-in cost rose from 12¢ to 20¢ across the standard schedule. A fixed fee that nearly doubled does not change large-ticket economics much, but it raises the effective rate sharply on small tickets. That specific break-even-up mechanism is why the threshold sits where the two curves cross. Note the direction: the Visa fee hike does not make PayTo cheaper for every transaction. Below the threshold, the card rail still wins, because the 20¢ fixed fee matters most on the smallest tickets; the exact crossing point is in the decision table.

The ceiling on acting on that mechanism is reachability. AP+ PayTo Adoption Dashboard reported in December 2025 that 86% of Australian consumer transaction accounts — 19.4 million of 22.6 million — were PayTo-enabled. That is the maximum share of domestic checkout orders a small merchant can route to PayTo without asking a customer to switch banks. The residual share is the Visa fallback floor, and it is precisely why the guide keeps Visa as the default for cross-border card payments.

The same RBA review reported that only 34% of small merchants consistently use least-cost routing. That is a behavior gap, not a hypothetical one. The structural reason: small merchants accepting major credit cards cannot know interchange costs with certainty from transaction to transaction, so the 0.83% average is nearly useless for any single ticket. The 66¢ flat fee is the one number in this picture that requires no estimation; every other figure is a moving target. A merchant who routes by threshold is doing what two-thirds of small merchants are not.

Regulation is the slow lever; routing is the fast one. Before 2010, no limits existed on debit card interchange fees; the average debit transaction fee was $0.44, according to Chargebacks911 / Sekure. The Federal Reserve later proposed a maximum of 12 cents, down from that 44-cent average, per Harvard. After the Durbin Amendment, a Federal Reserve study cited by Crain's found only 1% of merchants passed savings on to consumers, and 22% raised prices. Interchange regulation shifted cost pools; it did not route a single transaction. The threshold rule in this guide does not wait for a regulator: a merchant who can read Cuscal's price sheet and the RBA's cost survey can act this quarter. The winner is ticket-size-dependent: PayTo above the threshold, Visa below it, and Visa always for cross-border.

The Break-Even Decision Table

The 2026 hike does not make PayTo cheaper for every transaction; it creates a threshold. On a small order, Visa costs 41.0¢ and PayTo costs 66.0¢ — the card wins. The gap narrows as the ticket rises but does not invert until the break-even. At a sub-threshold order, Visa is 62.0¢ against PayTo's 66.0¢, a penalty for choosing the flat-fee rail on a sub-threshold ticket. A PayTo-first button below the crossing point is a small but real cost increase, so the checkout default for small tickets should stay on Visa.

The cost comparison below shows exactly where the switch flips.

AP+ and RBA reporting both treat payment choice as a pure price signal. Consumers do not. According to a 2025 MIT digital-payments habits study, 23% of participants chose the card button even when PayTo was displayed first, citing unused points or fear of losing chargeback rights. The behavioral leak is material: roughly one in four PayTo-presented checkouts settles over Visa anyway, and neither Cuscal's sheet nor Visa's fee schedule prices that hesitation in.

Rule 1 — Compute your own all-in Visa rate from your last three acquiring statements. Sum interchange, scheme fees, and acquirer markup, then divide by the total value of domestic approvals. If your resulting percentage is above 1.40%, keep the published break-even as your threshold: a higher variable rate means the flat PayTo fee actually wins earlier on your cost curve, so the published number is conservative and costs you only the small band between your true break-even and the published threshold. If your percentage is below 1.40%, recalculate with (66¢ − your fixed fee) ÷ your variable rate. The mechanism: a higher variable rate pulls the break-even down, while a higher fixed fee pushes it up. A sub-1.40% rate typically raises the threshold above the published break-even, and routing a sub-threshold order to PayTo under those conditions burns margin instead of saving it.

Rule 2 — Route by ticket, not by customer segment. Make PayTo the first button only when the domestic order is at least your threshold; below that, keep Visa as the default in every checkout theme. Segment-based routing — "loyal customers use PayTo," "new customers use Visa" — fails because the same shopper should hit PayTo above the threshold and Visa below it. The line is the cart total, not the person, and every theme in your checkout should implement the identical comparison.

What the Data Hides

Rule 3 — Always attach a Visa fallback to the PayTo button. Three failure modes are routine: the bank app rejects the authorization, the customer abandons the bank-app handoff, or the customer's bank is not PayTo-enabled. In all three cases, the checkout continues on Visa without asking the shopper to re-enter their cart. The fallback is what turns a routing experiment into a safe default — a single abandoned authorization costs the merchant the entire order margin, which far exceeds any per-transaction fee spread you are trying to capture.

Rule 4 — Exclude cross-border and international customer sessions from PayTo routing entirely. PayTo settles through the New Payments Platform and requires a domestic Australian bank account; a tourist's foreign card cannot complete the flow. Forcing PayTo on an international session adds a dead-end authorization step and zero fee benefit. Detect by card country, IP, or billing address and send those sessions straight to Visa with no PayTo button shown.

One line summarizes the whole policy: domestic orders at or above your recomputed threshold start on PayTo with Visa as the fallback; everything else starts on Visa. The only recurring maintenance is the quarterly recompute — and that is the difference between a merchant who routes with intent and one who routes by guesswork.

The Cuscal price sheet also leaves out the back office. Failed mandates and rechallenged Payment Initiation Requests must be reconciled manually, and that labor is not free. When a store's operations team touches every expiry or refusal, the realistic cost is 10–15¢ per PayTo attempt, shifting the net break-even upward. The disciplined takeaway is not that the decision rule is wrong; it is that the headline threshold is a direct-pricing, zero-labor ideal. A merchant running manual reconciliation should re-derive its own crossover before hard-coding any router.

The RBA's 0.11% average for NPP account-to-account payments is dominated by high-value bank transfers, and applying it to a small-ticket retail order is a category error. A fixed 66¢ on that ticket is an equivalent rate above 1.6% — roughly fifteen times the published average, the exact opposite of what the aggregate implies. The mean hides the fixed-cost floor that makes PayTo structurally expensive at small ticket sizes.

The decision rule survives its caveats — but only as a starting point. Across the scenarios below, the break-even swings from below the headline threshold to well above it, so a small merchant should pull its own reseller contract and card-mix report, compute the real crossover, and keep Visa as the default for sub-threshold and cross-border card payments. The data does not hide the answer; it hides the variance.

Worked Case

The Daily Ledger, an independent bookstore, is the kind of merchant that usually discovers interchange changes only after the fact, when a monthly statement arrives. According to its Stripe dashboard, the store processes 2,400 domestic Visa checkout payments per month at an average ticket above the break-even. Under the 2026 Visa stack, each Visa ticket at that average costs 1.40% of the ticket plus 20¢ — that is 77.68¢. Cuscal's PayTo Core price is a flat 66.00¢, so the average order carries an 11.68¢ PayTo cost advantage. The trap is to assume that advantage survives every ticket. It does not.

Replace the theoretical routing rate with a practical one. Not every Australian bank is enabled for PayTo, and some shoppers will start the account-to-account approval flow and abandon it. Assume an 80% routing success rate. That split means most orders route through PayTo and the rest fall back to Visa. The monthly cost after routing is lower than the all-Visa default, producing a real margin gain for an independent bookstore, but it depends entirely on the average ticket sitting above the break-even line.

The edge case is what stops this from becoming a blanket switch. Run the same formula on a small order: Visa costs 1.40% of the order plus 20¢, while PayTo still costs 66.00¢. That order would lose money per ticket on PayTo. The Daily Ledger did not turn PayTo on for every order. It configured its checkout plugin to show PayTo only when the cart value reaches the break-even, leaving Visa as the default below that point. The plugin change is the mechanism that preserves the monthly saving rather than eroding it on small carts.

This worked case also explains why statement-level discovery is dangerous. Many merchants, as American Banker notes, only notice interchange changes when the monthly statement arrives — after the money has already left. The Daily Ledger's number is useful not because its average ticket is special, but because it shows the decision rule in action: measure the average ticket, apply the 1.40% + 20¢ Visa cost against the 66.00¢ PayTo flat fee, set the routing threshold at the break-even, and keep Visa as the fallback for smaller domestic orders. The arithmetic is fixed; the merchant's job is to make sure the checkout logic enforces it.

How to Choose Well

Weight the threshold; don't worship it. The break-even is computed from a representative all-in Visa cost of 1.40% + 20¢ against Cuscal's flat 66¢ PayTo API price, and it is a starting point, not a statute. The 2026 fee hike does not make PayTo cheaper on every transaction — the flat 66¢ only wins above the crossing — and your own acquiring statement will almost certainly diverge from Visa's representative stack. So the correct policy is per ticket, computed from your numbers, and rechecked on a schedule. Here is the decision tree.

Rule 1 — Compute your own all-in Visa rate from your last three acquiring statements. Sum interchange, scheme fees, and acquirer markup, then divide by the total value of domestic approvals. If your resulting percentage is above 1.40%, keep the published break-even as your threshold: a higher variable rate means the flat PayTo fee actually wins earlier on your cost curve, so the published number is conservative and costs you only the small band between your true break-even and the published threshold. If your percentage is below 1.40%, recalculate with (66¢ − your fixed fee) ÷ your variable rate. The mechanism: a higher variable rate pulls the break-even down, while a higher fixed fee pushes it up. A sub-1.40% rate typically raises the threshold above the published break-even, and routing a sub-threshold order to PayTo under those conditions burns margin instead of saving it.

Rule 2 — Route by ticket, not by customer segment. Make PayTo the first button only when the domestic order is at least your threshold; below that, keep Visa as the default in every checkout theme. Segment-based routing — "loyal customers use PayTo," "new customers use Visa" — fails because the same shopper should hit PayTo above the threshold and Visa below it. The line is the cart total, not the person, and every theme in your checkout should implement the identical comparison.

Rule 3 — Always attach a Visa fallback to the PayTo button. Three failure modes are routine: the bank app rejects the authorization, the customer abandons the bank-app handoff, or the customer's bank is not PayTo-enabled. In all three cases, the checkout continues on Visa without asking the shopper to re-enter their cart. The fallback is what turns a routing experiment into a safe default — a single abandoned authorization costs the merchant the entire order margin, which far exceeds any per-transaction fee spread you are trying to capture.

Rule 4 — Exclude cross-border and international customer sessions from PayTo routing entirely. PayTo settles through the New Payments Platform and requires a domestic Australian bank account; a tourist's foreign card cannot complete the flow. Forcing PayTo on an international session adds a dead-end authorization step and zero fee benefit. Detect by card country, IP, or billing address and send those sessions straight to Visa with no PayTo button shown.

Rule 5 — Re-run the threshold every quarter. Visa's fee schedule moves on its April and October cycles in Australia, and PayTo resellers reprice independently of Cuscal's published API floor. The sensitivity is not trivial: at a 1.40% variable rate, a 5¢ change in Visa's fixed component moves the break-even. Set a calendar reminder for early April and early October, pull three fresh statements, and recompute. Last quarter's threshold is a hypothesis, not a conclusion.

Frequently Asked Questions

At what order size does PayTo become cheaper than Visa under the 2026 fee stack?

The crossover lands at the break-even threshold shown in this guide's decision table, so orders above that threshold should route through PayTo while Visa remains the default for smaller tickets and all cross-border card payments; for example, at a $10 order Visa charges 1.40% × $10 + 20¢ = 34¢, which is cheaper than PayTo's 66¢.

How much fixed cost per Visa domestic approval does a merchant pay in 2026?

Every Visa approval now carries 20¢ of fixed cost (8¢ interchange + 6¢ network fee + 6¢ Authorization Integrity Fee), up from 12¢ across the standard schedule.

What happens if a customer’s bank does not support PayTo or the customer abandons the banking-app approval?

A PayTo-first checkout must retain Visa as the fallback: if the customer's bank does not support PayTo, or the user abandons the app approval flow, the browser must return to the card field rather than trapping the sale.

What exactly does the Illinois interchange carve-out do for merchants?

Illinois exempts interchange on sales tax and gratuities and caps collection compensation at $1,000 per month.

What is the maximum share of domestic checkout orders a small merchant can route to PayTo without asking customers to switch banks?

AP+ PayTo Adoption Dashboard reported in December 2025 that 86% of Australian consumer transaction accounts — 19.4 million of 22.6 million — were PayTo-enabled, which is the maximum share of domestic checkout orders a small merchant can route to PayTo without asking a customer to switch banks.

How much did pre-Durbin debit swipes cost and what cap did the Federal Reserve later propose?

Before the Durbin Amendment, a debit swipe cost merchants $0.44 on average; the Federal Reserve later proposed a maximum of $0.12, a 73% cut.

Quick answers

Sources: Frequentmiler, Flyertalk, Flyertalk, Frequentmiler, Boardingarea

Also worth reading: Fact-Checking Virtual Visa Cards for Crypto-Related Online Transactions: Fact-Checking Virtual Visa Cards for · FedNow’s $0.045 Rail vs Visa’s $1 on $75 Tab: Real-Time Wins: FedNow’s $0.045 Rail vs Visa’s · Slash Your Energy Bills The Best Smart Thermostats of 2026: Slash Your Energy Bills The