Curaçao Reform vs Anjouan 2026: iGaming Payment Access Map

The Curaçao reform that ran from December 2024 into May 2026 changed who issues the license, but operators learned the harder change sat on the cashier side. Most published guides cover the license application; this article maps what comes after. Which tier-1 acquirers actually approve a post-LOK CGA MID in 2026, the orchestration plus crypto-native stack Curaçao books really run on, and where Anjouan stops looking like a substitute the moment the same operator tries to plug it into the same cashier.

Twelve providers in our database touch this map at production volume. Four card-side names with published Curaçao support and very different underwriting realities. Three orchestrators and a LATAM-specialist acquirer that carry CGA volume the tier-1 pool does not write directly. Four crypto-native gateways split by KYT posture. The combinations that actually run come out of those twelve.

December 2024 to December 2026: the LOK timeline acquirers read

The reform sequence the acquirer underwriter actually reads is six dated events, not one. Each event shifted a specific input in the underwriter's file: the regulator's identity, the available license stock, the principal screen, the deadline pressure on legacy operators, and finally the rejection-rate signal that priced where the new regime sat against MGA and UKGC.

An earlier version of this article carried an 87-of-140 application figure for the CGA's screen. The number does not trace to any official CGA publication, so it is removed; the CGA does not publish a running approval rate. What is verifiable is the shape of the screen. Per Coincub's LOK regime guide, the dominant rejection reasons are weak UBO disclosure, thin source-of-funds documentation, and links to high-risk jurisdictions, all of which mirror the same screens an acquirer's KYB pipeline runs. That is the read that matters at the underwriting desk: the CGA's gate is strict enough to have teeth, and nothing in the public record since the reform has pushed acquirers toward a vertical-wide Curaçao pull.

FATF and EU posture round out the underwriter's read. Curaçao is not on the FATF grey list as of mid-2025, despite occasional industry references to the contrary. The CFATF mutual evaluation report published in July 2025 placed Curaçao in enhanced follow-up rather than on the grey list itself. The EU removed Curaçao from its tax grey list earlier in the cycle. Neither change reopens the acquirer's risk model on its own; both close common reasons the model would reject.

The reform's net effect on payment access is one-sided. The CGA's screen tightened the supply of new Curaçao licensees, which acquirers welcomed in principle. Supply tightening is a slow input to acquirer-side underwriting models that still price the broader Curaçao book against five years of master-license-era TC40s and MATCH listings. The post-LOK acquirer roster has updated its language around Curaçao. It has not updated its pricing.

The tier-1 acquirer roster: published support vs MID reality

Four tier-1 names publish iGaming card acquiring at production volume. Their public license rosters cover Curaçao differently, and their actual underwriting positions on a CGA-direct kit in May 2026 differ further again. The table below pulls the score, category, settlement, and deposit-fee fields from the database; the prose underneath is the underwriting-side read.

Worldpay publishes Curaçao alongside FCA, UKGC, and MGA on its supported-license roster. The 2026 reality on a CGA-direct application is a 4 to 6 week onboarding when the kit is clean, the principals carry no MATCH residue from the master-license era, and the trailing six-month chargeback ratio sits below the acquirer's internal cutoff for the vertical. The friction concentrates on principals: a director with a current MATCH listing under a master-license-era termination kills the file before the chargeback math runs.

On Nuvei, the differentiator over Worldpay on Curaçao production volume is iGaming-specific risk tooling plus an acquirer pool that includes second-tier card acquirers under one orchestration roof. Nuvei's roster lists MGA, UKGC, Curaçao, Isle of Man, and US states. Operators with $2 million to $10 million monthly card volume and a CGA-direct license describe Nuvei as the production option more often than Worldpay, primarily because the orchestration layer absorbs the per-acquirer underwriting variance the operator would otherwise face directly. Onboarding times in the 2 to 4 week band are realistic for clean kits.

Adyen is the only one of the four whose published gambling roster never mentions Curaçao: it runs MGA, UKGC and a US, Australia, Canada and Brazil regulated set. Its 2026 underwriting position is also the most conservative of the four for CGA-direct applications. Operators who applied with a CGA-only kit report Adyen will quote Curaçao volume but won't underwrite without a recognized second license such as MGA or UKGC also on the file. The reasoning matches Adyen's pattern on every high-risk vertical: a license stack with multiple regulatory anchors prices better in Adyen's risk model than a single offshore one. In practice Adyen is a tier-1 default for dual-licensed CGA plus MGA books, not a path for CGA-only operators.

Where Paysafe sits is different again. Paysafe's published license roster does name Curaçao, alongside MGA, UKGC, Alderney and a US state set, and the gap between that roster and what it will actually underwrite is the widest of the four. What Paysafe runs in 2026 is case-by-case underwriting on post-LOK CGA applicants where the operator's deposit mix already routes meaningfully through Paysafe's e-wallet brands Skrill, Neteller, and paysafecard. The card underwriting becomes a value-add to an existing wallet relationship rather than the primary onboarding rail. The fit is European-licensed CGA operators with player bases that already use those wallets, not new applicants without prior Paysafe history.

Across the four names, published support is not the same as production MID issuance. CGA-direct applicants without a second recognized license on the kit face a tier-1 acquirer pool that quotes politely and underwrites narrowly. Tier-1 acceptance on CGA-only books in 2026 is the exception, not the norm, and that gap is what the rest of the stack exists to close.

The CGA stack in production: orchestrators plus crypto-native rails

For CGA-only operators, or for any CGA-licensed book where the tier-1 path stalls on principals or chargeback math, the production stack in mid-2026 looks different. An orchestration layer routes to specialist tier-2 acquirers that price Curaçao volume to their own internal risk model, and a crypto-native rail handles the segments where card approval is structurally poor.

Two orchestrators carry credible Curaçao routing in 2026.

IXOPAY, a TokenEx company since their April 2024 merger, runs enterprise high-risk vaulting plus tokenization under a single API. The orchestration fee sits at 0.1 to 0.5 percent on top of the underlying acquirer cost, with the value being failover when one acquirer in the pool tightens or pulls Curaçao support. IXOPAY does not underwrite directly; it routes to acquirers who do, which is exactly the architectural question that matters when one of those acquirers exits the vertical mid-contract.

Where Finera differs is breadth of routing. The Cyprus-based orchestration platform connects 600+ payment providers behind one API with AI-driven smart routing, positioned for high-risk and iGaming as named verticals. The fit is Curaçao operators who want a single technical integration that abstracts whichever specialist tier-2 acquirer is currently underwriting CGA volume at acceptable terms. The orchestration layer recovers approval rate that the operator would lose by going direct to one acquirer and hitting a per-acquirer ceiling.

PayRetailers occupies a different position in the stack. It is not an orchestration layer; it is a specialist acquirer whose published roster covers Curaçao, MGA, and LATAM markets through seven local licenses across Latin America. For CGA operators with player books concentrated in Brazil, Argentina, Mexico, or Chile, PayRetailers is often the underlying acquirer the orchestrator routes to, not a substitute for the orchestrator. The 300+ local payment methods PayRetailers aggregates across LATAM are also the methods that recover funnel completion the card rail cannot cover in those geographies.

On the crypto-native side, the Curaçao question is KYT posture, not coin breadth.

CoinsPaid was one of the dominant rails for Curaçao volume during the master-license era and carried that book into the LOK regime, but its EU footing changed at the July 1, 2026 MiCA cliff: the Estonian registration voided, a replacement CASP application is under review, and the company runs a supervised restricted mode with no new merchant onboarding while it waits. The MiCA register audit tracks that status. For a CGA book this is a counterparty question rather than a legality one, since Curaçao sits outside the EU perimeter, but a new stack cannot onboard onto it today. It is a custodial gateway: player crypto lands in CoinsPaid-controlled wallets before settlement, and its July 2023 hot-wallet breach (more than $30 million, attributed to the Lazarus Group) is the concrete shape of that risk. What existing CGA books bought with the custody trade-off is the deepest iGaming-specific feature set among the crypto rails here. Independent reporting puts Bitcasino, King Billy, and PlayAmo on its customer roster.

CoinGate publishes MiCA, MGA, and Curaçao on its license roster. It runs Chainalysis-grade KYT on inbound deposits under its MiCA CASP authorization from the Bank of Lithuania, and the chargeback liability sits at zero by structure rather than by negotiation. The fit is Curaçao operators who want EU-regulated crypto compliance without putting the LOK ordinance's mandatory blockchain-analytics burden entirely on the operator's own stack.

NOWPayments is the breadth option at 350+ supported assets and 0.5 to 1 percent fees. Its published license roster is one of only two in our database that explicitly list both Curaçao and Anjouan, alongside Praxis Tech, which makes NOWPayments the cross-license crypto rail for operators who hold both. It holds no EU authorization of any kind, which is immaterial for a pure CGA book and disqualifying the moment the same group runs an EU-licensed entity on the same rail. The custody posture is the inverse of CoinsPaid's: non-custodial by default, with payments forwarding to the operator's own wallet as they clear, and a custodial flow available for operators who prefer platform-managed conversion.

CoinPayments is the documentation-light option, with a coin list that shrank to roughly 40 assets after its platform revamp. Production fit on a Curaçao book is as a secondary rail behind one of the three already named, picking up the segments those gateways won't service on compliance grounds.

The running CGA stack in mid-2026 is one orchestrator plus one or two crypto rails. The orchestrator carries the card volume tier-1 acquirers won't underwrite directly; the crypto rails carry the segments where card approval is structurally weak. Tier-1 names enter the picture only when the operator's license stack adds MGA or UKGC alongside CGA, at which point the article is no longer about Curaçao-only economics.

Anjouan head-to-head: where AOFA paperwork dies at KYB

Anjouan is registrable as a license. The merchant-account path is where the head-to-head with Curaçao stops being head-to-head, and the reason runs through three statements that don't appear on the marketing pages of license intermediaries.

First: the Banque Centrale des Comores has stated publicly across communiqués in 2022 and 2023 that no island entity, including the Anjouan Offshore Finance Authority, is authorized to issue or supervise financial-sector licenses. Per Legasset's reading of the BCC notices, the relevant language is "No island entity, including the so-called 'Anjouan Offshore Finance Authority', is authorised to issue or supervise banking or financial licenses." That language is the input third-party screening tools use during automated KYB.

Second: the FATF declined to assess the Comorian gambling sector in its 2024 mutual evaluation because gambling is explicitly prohibited under the Comorian Penal Code, per Tradepass's analysis of the relevant FATF report. So even if AOFA paperwork passed the BCC's authority test, the underlying jurisdiction's penal law would surface a separate flag at the acquirer's compliance review.

Third: the operational consequence. Applications citing AOFA paperwork either get refused at intake by automated KYB pipelines, or escalate into enhanced due diligence where the underwriter asks for an alternative license from a recognized regulator before the file moves further. The 2026 outcome is that Anjouan-only books have no card-acquiring path through any tier-1 or tier-2 acquirer with public iGaming exposure. Specialist offshore PSPs that do quote Anjouan operators charge premium pricing, hold reserves at higher percentages, and shorten contract terms relative to the Curaçao equivalent.

Crypto-native is where Anjouan does have a path. NOWPayments accepts Anjouan-licensed operators on its published roster. CoinPayments and lighter-compliance crypto gateways serve Anjouan volume in production. The economics work when the operator's player base is crypto-native to begin with and the cashier doesn't need card support. SOFTSWISS markets an Anjouan turnkey package at €17,000 application and €13,300 annual renewal, with explicit messaging that Anjouan is positioned for crypto-first operations rather than traditional card acquiring.

The honest read on Anjouan in 2026 is that it is not a payment-access substitute for Curaçao. It is a fast and cheap entity wrapper for crypto-only segments of a multi-license operator, or a 2 to 4 week launch license for a book that will migrate to a recognized regulator once that license issues. Per Altenar's comparison, the cost gap to a CGA license is meaningful only if your model already runs without PSP acceptance from mainstream providers. The moment cards enter the cashier, the gap closes the wrong way.

Holding both licenses: when CGA plus Anjouan changes the math

The dual-license question is genuine for two operator profiles.

The first is a Curaçao-licensed operator whose CGA license restricts geographies the operator wants to reach. The CGA's compliance regime under LOK has tightened territorial restrictions on what a CGA-direct license is permitted to cover. Anjouan, silent on most jurisdictions, lets the operator structure a separate entity for traffic the CGA-direct license cannot legally support, with the cashier on the second entity running entirely through crypto-native rails. The card stack stays attached to the CGA entity; the Anjouan entity becomes a routing question for crypto-only volume the parent group decided not to refuse.

The second is a new entrant priced out of the LOK fee ladder and timeline. Anjouan's 2 to 4 week issuance and €17,000 application cost provide a launch license while a CGA application goes through the CGA's two-phase process at roughly 8 weeks per phase. The cashier on the Anjouan entity is crypto-only by design. The operator's intent is to migrate the book to the CGA entity once that license issues, with the Anjouan entity shut down or repurposed for crypto-only segments.

What dual-licensing does not do: it does not extend tier-1 card-acquiring access to Anjouan-only volume routed through the second entity. The acquirer's KYB read of the underlying entity is what gates the MID; the parent group's other licenses don't transfer. Operators occasionally try to consolidate Anjouan player traffic into the CGA MID's card processing flow. The underwriter's transaction-monitoring read surfaces the geo-mismatch within the trailing six-month window and the MID gets pulled.

The decision rule reduces to one sentence. If the cashier needs cards, the license you submit to the acquirer has to anchor the MID, and CGA does that while Anjouan does not. A dual-license stack where Anjouan adds something does so in segments where the cashier is crypto-first by design. In every other segment, Anjouan is paperwork that doesn't change the cashier and a recurring fee that doesn't earn its keep. Pick CGA when the operator's medium-term plan includes card acquiring; pick Anjouan only when it doesn't, and keep the cashier on crypto-native rails matched to that choice.