Consumers Can Complain to AFCA About Chargebacks. Merchants Can't. Should They Be Able To?

The short answer from the data is no. Not because merchants don't deserve better, but because the fix isn't giving them the same process. What the data does show clearly is that when banks get a chargeback process — including their ePayments Code obligations — wrong, it matters, and it needs to be caught and corrected. The harder question is how you build that accountability without turning every dispute into a case-by-case argument about facts nobody can actually prove.

PDI reviewed all AFCA cases lodged between 14 June 2024 and 2 August 2026 under the categories "chargeback – declined (consumer)" and "chargeback – delayed." Over that period there were 65 such complaints: 35 in 2025, 13 so far in 2026, and the remainder in the partial 2024 window. Some of these were scam bank transfers, fee disputes, and insurance cases captured under the same category label rather than genuine chargeback disputes. Excluding these, the net chargeback-related scenarios come to 54.

Of those 54, only 3 were won by the complainant. That's 94.4% won by financial firms.

This is only the data that's published online. The bigger question is what the impact on chargebacks is when the threat of an AFCA complaint is hanging over a financial institution. There is no data on how many consumers are telling financial firms "raise a chargeback / give me my money back or I tell AFCA," or on how a potential AFCA threat is shaping standard operating procedure inside disputes teams. If a threat and pressure can only come from the consumer's side, that naturally creates bias.

Two of the three complainant wins were clear ePayments Code and scheme rule violations. In one, a transaction was deemed unauthorised, yet the Code wasn't followed. In the other, communication timeframes and applying the correct reason codes weren't adhered to. That's a straightforward compliance failure, not an argument about outcomes. It's worth noting that many banks proactively self-report to ASIC when they identify breaches — for example, system issues that cause communication timelines to lapse.

Worth flagging: the ePayments Code is currently a voluntary industry code. Banks choose to subscribe to it, but once they do, compliance becomes a contractual term with their customers — which is how these two cases could still be enforced. That may be about to change, though it isn't law yet. Treasury has drafted legislation that would let the Minister make the ePayments Code mandatory for everyone, rather than leaving it as something banks opt into. That draft law was out for public consultation earlier in 2026 and hasn't passed Parliament yet, but the direction of travel is clear: turn a voluntary code into a binding one, set by government rather than left to individual banks signing up. That shift matters for where this piece ends up.

The third case is a more ambiguous scenario for backoffices to work through: a financial firm accepting insufficient merchant evidence.

A $2,513.92 dress was disputed as "goods and services not as described" — the consumer said a patch of sequins was missing. The merchant provided a range of evidence: delivery confirmation, their terms and conditions, and records of customer communication. What they didn't have was any evidence that could confirm all the sequins were sewn on in the first place — and how do you prove that? AFCA ruled in favour of the consumer. Read our deep dive into the evidence problem space: The Evidence Problem: Why Dispute Evidence Is Both Overloaded and Undervalued

What we do know is that the financial firm faced a large bill for a consumer-merchant issue. AFCA found the firm had "accepted the merchant's response at face value without considering whether the merchant's claims could be supported," and that had it met its obligations, the chargeback would have been resolved in the complainant's favour. The firm paid for the dress plus interest, paid $500 to compensate for non-financial loss, paid AFCA's fee when the case was lodged, and paid again when AFCA ruled against it. The total cost of a disputed dress would run into the tens of thousands.

It's worth asking: is a consumer-facing complaints process the right way to handle this, given that in near-95% of reported and published claims, financial firms are found to be doing the right thing from a process perspective? The majority of chargeback-related cases are consumers unhappy about an outcome — not a process.

AFCA can require a financial firm to pay a complainant for non-financial loss, up to a maximum of $6,300 per claim, even where the underlying chargeback decision is upheld. Across the determinations reviewed, there were 6 instances of non-financial loss awards of up to $1,000 — for inconvenience, stress, and anxiety caused, and for having to repeatedly provide information the consumer had already provided (hello, generic "we need more information" emails).

As a consumer, this is not a bad business case: an 11% success rate on a non-financial loss payout, in this small sample, at no cost to lodge.

Providing consumers a pathway to escalate matters is genuinely important, and an important legacy of the Royal Commission. There are still issues with financial firms meeting their obligations, and even at under 5%, we need oversight and accountability for that.

But the instinct to fix the asymmetry by extending the same mechanism to merchants doesn't hold up. AFCA's current model detects process failures by re-litigating the underlying factual dispute case by case — was there really a missing sequin, was the item really not as described, did the delivery really not arrive, was a transaction really unauthorised. Open that same door for merchants and you don't get accountability, you get volume and chaos: a court of missing sequins, run in both directions, for every "not as described," every "not received," every ambiguous item across every merchant category. It doesn't scale, and — as the dress case shows — it often doesn't actually settle anything. Nobody will ever know if the sequins were there.

What the industry needs instead is a rules layer, not a court. A black-and-white application layer of chargeback rights, sitting alongside Australian Consumer Law — publicly available, able to be referred to by banks, merchants, and consumers alike, open to industry feedback, and ratified by a body like AFCA or an ombudsman so it actually carries weight. That reframes what a body like AFCA is for in this context: not the forum that re-decides whether the sequins existed, but the body that ratifies and holds the industry accountable to the rules everyone is supposed to be applying.

This isn't a hypothetical model. It's already being proposed one step over: draft legislation currently working its way toward Parliament would move the ePayments Code from a voluntary code to a mandatory one, set by the Minister rather than left to individual banks choosing to subscribe. If unauthorised-transaction rules are heading toward being codified and made mandatory rather than tested case by case through AFCA, the same logic should apply to chargeback rights.

Detection and enforcement of process compliance can then happen the way it does in other regulated industries — through direct testing rather than a queue of individual complaints. AFCA-run spot checks — mystery disputers, in the same spirit as mystery shoppers — using structured test scenarios across the handful of dispute channels and the handful of banks that actually exist in Australia, would test whether a firm's process holds up against the ePayments Code and the rules layer directly. That's the same idea as a product or change manager testing a process before it ships, rather than waiting for enough individually aggrieved consumers to surface the same failure one case at a time.

Scams are a different problem, and they're often tangled up with chargebacks in practice — but they're not the same thing. A chargeback is only a recovery mechanism for funds already lost. A consumer-facing complaints process remains crucial for scams specifically, because it's one of the few channels through which patterns of buying and selling scams get detected and disrupted early. The case for replacing case-by-case adjudication with a rules layer applies to chargebacks. It doesn't apply to scam detection, where the value of a consumer-facing pathway is different in kind.

Chargebacks currently don't have a clear regulatory home. AFCA tests process. Schemes own the rules. ASIC oversees the ePayments Code. The RBA holds the broader payments policy mandate. AusPayNet, the industry's self-regulatory body, already owns existing code sets, and its Issuers and Acquirers Community does include large merchants like Coles and Woolworths alongside the banks — though its structure remains bank-founded. None of them is currently positioned to convene the kind of independently ratified rules layer this piece is arguing for. Which of these — or which new body — would actually take that on is the first open question.

The second is what it would take for that conversation to even start, especially while chargebacks continue to sit in the backseat of payments policy discussions — even as Mastercard's 2025 State of Chargebacks Report (with Datos Insights) forecasts chargeback volume in Asia Pacific will grow 35% between 2025 and 2028, from 35.1 million to 47.4 million disputes a year, and first-party abuse has become a major issue for merchants and issuers alike.

A narrower thread also sits outside the scope of this piece and belongs on PDI's roadmap: "financial firms," as AFCA's own data groups them, is a wide umbrella that includes staged digital wallets and BNPL providers. Those firms are typically measured against their own user agreements and proprietary buyer protection policies rather than standard chargeback rules — but a consumer can also raise a chargeback against those same firms through their issuing bank, meaning two different, overlapping pathways can apply to the same transaction. How that works in practice, and what it means for consumers and merchants, needs its own analysis.

Full breakdown:

This table reflects all 65 raw complaints under both AFCA categories, before excluding the 11 non-chargeback matters (scam bank transfers, insurance, fee disputes) that sit inside the same category label — it is not the same 54-case base used for the win/loss figures above.

This article discusses and analyses published, de-identified determinations made by the Australian Financial Complaints Authority (AFCA). AFCA is acknowledged as the maker and author of the determinations referenced in this piece. AFCA's website can be found at www.afca.org.au. Links to the specific determinations discussed in this article:

No content from these determinations has been altered. Where determination text is referenced in this article, it has been paraphrased, or quoted briefly and clearly marked as a quotation, with attribution to AFCA. This article does not reproduce any determination in full, and nothing in it is inconsistent with the de-identification of the determinations it discusses.

The analysis, commentary, and views expressed in this article are those of the Payment Disputes Institute (PDI) and its author, and are not those of AFCA. This distinction applies throughout the article: any statement of fact drawn directly from a determination is AFCA's; any interpretation of what that fact means for the industry is PDI's.