South Korea Payments Operator Guide: Local Acquiring, PGs, and VANs

South Korea's payment market is built on a paradox: the most card-saturated consumer market on earth is also one of the hardest for a foreign operator to enter with a direct card-acceptance setup. Korean consumers carry multiple credit cards, use them constantly, and have structured their entire retail economy around card rewards and installment plans. Yet the infrastructure that enables this is deliberately domestic — cross-border acquiring isn't an option, and local acquiring is mandatory for Korean-issued cards, both online and offline.

For a foreign operator, this is the first thing to understand. Korean local acquiring runs through two separate licensed layers depending on the channel: a PG for online cards, a VAN for in-person terminal transactions. Which layer you need shapes every other decision — which intermediary you work with, whether you need a local entity, how much you pay in fees, and how you handle data residency. This guide covers how the system actually works.

The Most Card-Saturated Market on Earth

Korea's relationship with cards is structural, not accidental. A government-driven formalization push in the late 1990s — combined with deep tax incentive programs tied to card spending — produced a consumer population that genuinely prefers cards for everything, including small purchases that most markets handle with cash or QR.

The 2025 Bank of Korea data crystallizes the scale: daily-average card spend reached approximately KRW 3.1 trillion, of which approximately KRW 1.7 trillion flowed through mobile — up 7.3% year-on-year. Mobile's share of total card spend rose from 52.4% in 2024 to 54.3% in 2025, meaning mobile has overtaken physical cards as the larger half of Korean card volume (Bank of Korea, 2025). Smartphone penetration sits at approximately 95% (Korea Communications Commission, 2024).

Total card transaction value reached approximately KRW 1,255 trillion (~US$934 billion) in 2024, with credit and charge cards accounting for roughly 80% of value and debit approximately 20%.

Takeaway for operators: cards are the substrate. The wallets, installment plans, and simple-pay apps that layer on top of them are largely card-on-file overlays — not replacements.

The Local-PG Requirement (Online Acceptance)

This is the structural constraint that defines Korean payment access for foreign operators taking cards online. Korean-issued cards cannot be acquired by a foreign acquirer directly. Domestic-card acceptance requires routing through a licensed Korean payment gateway or acquirer. (For in-person, terminal-based acceptance, the equivalent local-acquiring layer is the VAN — covered separately below.)

Two acceptance paths exist for most foreign merchants:

Path 1 — Domestic Korean cards: route through a licensed Korean PG. The major players are Toss Payments, KG Inicis, NHN KCP, and NICE Payments. KG Inicis processes only Korea-issued cards; overseas-issued cards cannot run through it. Toss Payments, KG Inicis, NHN KCP, and NICE together hold an estimated majority of the domestic PG market (industry estimates; precise shares are not officially published).

Path 2 — Foreign-issued Visa/Mastercard and PayPal: a cross-border PG. Eximbay is Korea's leading cross-border PG; 2C2P and PayLetter serve specific verticals including content and gaming. This is often a separate integration from the domestic-card route.

Business Registration Number (BRN): contracting a Korean PG directly for a merchant account generally requires a BRN — which means a local legal entity (Korean company, FDI company, or branch). E-commerce businesses also typically need 통신판매업 (mail-order business) registration. The way around this for operators without a local entity is an aggregator that holds the local PG relationship on your behalf — Stripe, KOMOJU, and Eximbay all offer this model.

For a broader view of how South Korea's market compares with APAC peers, the payment structure context is on the market page. The local-PG requirement here is one instance of a broader pattern operators run into market after market — see the local acquiring vs cross-border operator reference for the reach/approval/economics/regulatory framework behind it.

The VAN Layer: Offline Acceptance and Terminal-Level Card Processing

Everything above describes the online-facing layer. Korea has a second, structurally distinct local-acquiring layer that most guides to the market skip entirely: the VAN, or Value-Added Network. If your business takes cards in person — a store, a kiosk, a restaurant, a pop-up — the PG relationship covered above does not get you there. You need a VAN.

What a VAN actually is. A VAN sits between card companies and physical merchants. It installs and connects the card terminal (CAT), POS system, or kiosk; transmits the card-approval request generated when a card is swiped, inserted, or tapped; and handles the network-level switching that gets that request to the right issuer and back. A PG, by contrast, is the online-facing aggregation layer for merchants who cannot or do not want to contract with card companies directly. Toss Payments — itself a licensed PG — draws the distinction plainly in its own merchant-facing documentation: VAN is the offline "pipe" for terminal transactions, while PG is the intermediary for online payments, and expanding from one channel to the other means acquiring the other layer, not extending the one you already have.

Who the significant operators are. VAN operators register with the FSC under the Specialized Credit Finance Business Act (여신전문금융업법) — mandatory since a July 2015 amendment, with minimum capital and staffing thresholds. NICE Information & Telecommunication is Korea's largest VAN by processed-transaction share — 27.8% of total VAN transaction volume in 2024, up from 20.7% in 2020 — and, like several peers, also runs a PG business alongside its VAN business as a separate line. KICC (한국정보통신) operates two explicitly separate platforms: EasyPAY for online PG and EasyCheck for offline VAN. KSNET, one of the largest VAN/PG dual operators, serves several hundred thousand merchants across card VAN, PG, and banking VAN. KIS Information & Communications is another established VAN operator. The pattern across all of them: VAN and PG are licensed and operated as distinct service lines, even inside a single company.

Which layer you actually contract with. For an in-store, terminal-based business, the merchant agreement is with a VAN (or a VAN-affiliated terminal reseller who installs the hardware on the VAN's behalf) — not a PG. For an online storefront, it's a PG. If you need both — a physical retail presence and an e-commerce storefront — you generally need two separate contracts and two separate integrations, unless you use a provider that bundles both service lines (several of the larger Korean operators, like NICE, KICC, and KSNET, do). Foreign-operator aggregators such as Stripe and KOMOJU, as covered above, solve the online/PG leg without a local entity — they do not solve offline VAN acceptance. PaymentBrief found no equivalent "no local entity" shortcut for terminal-based acceptance; a VAN merchant contract requires a Korean Business Registration Number, the same requirement that applies to a direct PG contract.

Where VAN economics sit relative to the regulated MDR tiers. This needs care, because the two layers are billed differently. A VAN operator invoices the card company, not the merchant, for its fee — a structural difference from a PG, which invoices the merchant directly. That VAN fee is then one of the named cost inputs — alongside funding cost and risk-management cost — in the 적격비용 (eligible-cost) methodology the FSC uses for its periodic re-appraisal of the regulated MDR tiers already covered above, under the same Specialized Credit Finance Business Act. In other words: for offline transactions, the VAN fee is folded into the cost base behind Layer 1's regulated tiers, rather than sitting as a separate fee stacked on top of them the way the FSS-disclosed ~1.98% PG/intermediary average sits on top of Layer 1 for online transactions. This site cannot confirm the exact weighting VAN costs carry in that methodology, or state a specific VAN fee percentage merchants should budget — PaymentBrief did not find the eligible-cost schedule's full itemization in primary form, and none is asserted here. What is sourced: how VAN-to-card-company billing works, and that VAN fees are a named input to the same regulated-fee methodology, not a separate charge layered onto it. Separately, how VAN operators are compensated changed structurally in 2018 — a flat per-transaction fee (roughly KRW 70–150 regardless of ticket size) was replaced with a rate-based fee (approximately 0.28% of transaction value) effective 31 July 2018, following an FSC/FSS announcement aimed at reducing the burden on small, high-frequency merchants like convenience stores and pharmacies, while raising VAN-related costs at high-ticket merchants.

Terminal and reconciliation consequences. For a foreign merchant with a Korean entity opening a physical location, the operational sequence is: register the business, then contract with a VAN (or VAN-affiliated terminal vendor) to get a card terminal installed and connected — a distinct process from the PG integration used for the same company's online storefront. On reconciliation: because a VAN transmits approval data to each card company individually, offline settlement runs per card company, on that card company's own settlement schedule, rather than through the single consolidated window a PG typically provides across payment methods. A foreign operator running both channels should expect to reconcile online sales through one consolidated PG statement and offline card sales through multiple, card-company-specific settlement timelines routed through the VAN relationship — not a single unified ledger by default.

Simple-Pay Wallets: Card-on-File, Not Displacement

Korea's major consumer-facing wallets — collectively called 간편결제 (simple pay) — are KakaoPay, Naver Pay, Toss (Viva Republica), Samsung Pay/Wallet, Payco, SSGPay, and Zero Pay.

The most important operator-facing point: most simple-pay wallets are card-on-file overlays. They tokenize underlying domestic cards and still settle through the card/PG rails. "Accepting KakaoPay and Naver Pay" typically comes bundled through your local PG — not as a separate acquiring relationship with Kakao or Naver directly.

For qualitative market positioning: Samsung Pay leads in-store (NFC/tap-to-pay); Naver Pay leads online via its native integration with Naver Smart Store and Naver Shopping; KakaoPay operates as a super-app within KakaoTalk (46M+ monthly active users) covering payments, insurance, brokerage, and credit scoring. Toss bundles payments with a digital bank and brokerage platform, with particularly strong engagement among younger Korean consumers. These positions are from industry reporting — Bank of Korea does not publish wallet-level market share.

The digital wallet layer in Korea is thus best understood as a UX and loyalty layer built on top of the card infrastructure, not a parallel acceptance network. The practical integration consequence: contract the right PG, and wallet acceptance comes with it.

KFTC, Open Banking, and the Won Rails

The Korea Financial Telecommunications & Clearings Institute (KFTC) operates Korea's interbank retail network, including the Open Banking hub, the interbank transfer system, and multiple clearing systems across the Korean financial sector.

Open Banking launched in December 2019 — making Korea one of the earlier movers on centralized open banking infrastructure in Asia. All major banks plus dozens of licensed fintechs connect via a KFTC-managed API hub. This is the backbone for A2A transfers, simple-pay wallet top-ups, and direct debit flows. The Open Banking API lets licensed operators access account information and initiate transfers across all participating banks via a single integration point. For operators, this matters most for virtual account-based checkout flows and B2B settlement.

BOK-Wire+, the Bank of Korea's RTGS system, handled a daily-average value of approximately KRW 617 trillion in 2024 (+11.2% YoY) — the high-value settlement layer that sits above the retail rails.

Virtual accounts remain significant for B2C e-commerce, particularly for higher-value transactions where consumers prefer bank-confirmed payment over a wallet charge. A unique per-order virtual account number is generated; customers pay via internet banking or ATM within a set window. Reconciliation is automatic.

For regional context on how Korea's real-time infrastructure compares with Singapore, Thailand, and Indonesia, see the Real-Time Payment Rails Comparison Matrix.

Card Fees: A Two-Layer Cost Structure

Korea's card merchant fees are government-influenced via the FSC — a system with no close parallel in other major APAC markets. Understanding the cost structure requires distinguishing two separate layers.

Layer 1 — FSC-regulated MDR tiers. Under the Specialized Credit Finance Business Act, the FSC conducts a cost-based fee re-appraisal approximately every three years and sets preferential rate caps by merchant annual sales bracket. The most recent reform (announced 24 December 2024, effective 14 August 2025) reduced:

• Preferential credit-card MDR: from 0.5–1.5% to 0.4–1.45% (by annual sales bracket)
• Preferential debit/check card MDR: to 0.15–1.15% (by annual sales bracket)

Approximately 95.7% of Korea's roughly 3.21 million merchants qualify for these preferential rates. These are the regulated merchant discount rates — the cap that card companies charge small and medium merchants.

Layer 2 — PG/e-finance intermediary fee. Separately, the FSS disclosed in September 2025 that the average card payment fee across 18 e-finance (PG/intermediary) firms was ~1.98% (prepaid: ~1.74%). This is the fee charged by the PG/intermediary layer — a distinct cost on top of the FSC-regulated card MDR, not a replacement for it.

For most foreign operators: the preferential MDR rates apply to small domestic merchants, not to large foreign operators. The FSS-disclosed ~1.98% intermediary average is the more operationally relevant reference point for what most foreign merchants will actually pay. Total blended cost = regulated MDR + PG/intermediary layer.

Halbu (할부) — installment payments. Korea's deeply embedded credit card installment system allows purchases above approximately KRW 50,000 to be split into 2, 3, 6, 12, or 24 monthly installments. Most consumer purchases above mid-ticket in categories like electronics, fashion, and travel are offered with interest-free halbu; the merchant subsidises the financing cost. Local PGs surface halbu automatically. Foreign operators who do not expose halbu at checkout in these categories will lose conversion.

Regulatory Framework: FSC, FSS, EFTA

The Financial Services Commission (FSC) sets financial-sector policy; the Financial Supervisory Service (FSS) handles supervision and compliance. Both govern the Electronic Financial Business licensing framework.

Governing statute: the Electronic Financial Transactions Act (EFTA) is the core law for electronic financial transactions. Operating as a domestic e-finance business requires an Electronic Financial Business licence — with categories including Payment Gateway Operator, Electronic Prepayment Means (stored value/e-money), Electronic Funds Transfer, and Electronic Money Issuer. A domestic entity is required for direct licensing in practice.

Authentication: the mandatory ActiveX/accredited-certificate regime was abolished on 10 December 2020 via the Digital Signature Act amendment, switching Korea to a technology-neutral authentication framework. In practice, App Card, ISP (Internaional Secure Payment), simple-pay PIN/biometrics, and 3DS coexist. The legacy ActiveX requirement is gone — but Korea-specific app-based authentication (App Card) remains the norm for card-not-present flows, which is a real UX and integration consideration for foreign checkout operators.

Data Residency: EFSR and the EU Adequacy Exclusion

This is an area where precision matters for operators designing system architecture.

Korean financial-sector rules (EFSR) require domestic storage of financial data. Public-cloud systems handling financial or credit information must sit on servers located in Korea. Network-segregation rules require separation of internal and external networks — the FSS has treated offshore access to financial systems (including "global support" access) as a potential violation. Phased relaxation has been underway since August 2024, but this is a gradual process, not an open door.

EU adequacy decision (September 2025): PIPC (Korea's data-protection authority) obtained an EU adequacy decision that eases KR→EU personal-data transfers. However, this decision expressly excludes personal credit information and FSC-supervised credit data. It does not automatically permit offshore processing of Korean financial data and does not override EFSR requirements. Foreign operators should not assume that the EU adequacy decision covers their financial-data architecture — it does not.

Practical implication: design for in-country financial-data storage from the outset. Confirm the scope of any cloud or data-processing arrangement with your Korean PG and legal counsel, particularly if you operate global infrastructure that would otherwise process Korean financial data outside Korea.

Cross-Border QR Interoperability

A note on QR in Korea: the wallet landscape is well-served by in-app simple pay, and there is no verified domestic unified national QR standard equivalent to Indonesia's QRIS.

What is verified: KFTC is leading cross-border QR interoperability work, with a Korea-Indonesia link reported live approximately April 2025 and links to India and Vietnam planned. This is an outbound connectivity story — enabling Korean wallets to be used by Koreans travelling overseas, and vice versa — not a domestic QR standardization initiative. Frame it accordingly.

FX, KRW, and Settlement Friction

KRW is the domestic settlement currency, and the won is not fully internationalized. FX is liberalized but not free-flowing.

Key operational realities for foreign operators:

• Settlement is in KRW domestically; cross-border merchants settle out via FX banks
• Repatriation of proceeds above approximately US$500,000 must generally occur within ~18 months; larger flows trigger declaration and documentation requirements under MOSF/BOK reporting rules
• Payout timing via aggregators runs approximately T+4 (US accounts) / T+7 (outside US) per Stripe Korea documentation
• KRW presentment — Korean consumers expect KRW at checkout; FX conversion at the acquirer side, not consumer side

Budget for an FX-bank relationship and longer payout timelines relative to, say, Singapore or Hong Kong. For operators expecting significant Korea revenue, capital controls and repatriation friction are worth building into treasury planning.

What This Means for Operators

Entry paths, ranked by speed and structural commitment

Path 1 — Aggregator, no local entity (fastest): Stripe supports Korea without a local entity via a local processor, accepting Korean domestic cards plus KakaoPay, Naver Pay, Samsung Pay, and PAYCO with KRW presentment (Stripe Korea docs, 2026). KOMOJU offers a similar aggregator model for Korea with Japanese-market integration. This path works well for most foreign merchants entering the market.

Path 2 — Cross-border PG: Eximbay is Korea's leading cross-border PG and handles international cards plus select domestic methods. Useful for predominantly international audiences transacting in Korean storefronts, or for gaming/content verticals (PayLetter).

Path 3 — Direct local PG contract (full domestic economics): Toss Payments, KG Inicis, NHN KCP, or NICE Payments. Full domestic-card economics, halbu integration, and complete wallet bundling. Requires a BRN (local entity). Necessary if you are the merchant of record in Korea or need the full local fee structure.

Non-obvious requirements

• Two-PG reality: a domestic PG (Korean-issued cards) and a cross-border PG (overseas-issued Visa/Mastercard, PayPal) are structurally separate. A single integration through Stripe or KOMOJU resolves this; a direct PG contract means managing both legs
• PG covers online only — offline needs a VAN: a physical, terminal-based presence (store, kiosk, restaurant) is acquired through a VAN (NICE, KICC, KSNET, and others), not the PG relationship used for e-commerce. Aggregators like Stripe don't reach this layer; expect a separate contract and separate settlement timeline
• Wallets are bundled, not standalone: KakaoPay, Naver Pay, and Samsung Pay come via the PG — not separate acquiring relationships
• MDR has two layers: regulated preferential MDR (small domestic merchants) and PG/intermediary fee (~1.98% FSS avg). Large foreign operators pay the latter, not the former
• Data residency is real: financial data stays on Korean servers. Design for this; the EU adequacy decision does not cover FSC-supervised financial data
• Halbu is expected: for mid-ticket and above in consumer electronics, fashion, and travel, exposing installment options at checkout is a conversion requirement, not a feature

Regulatory watch (2025–2026)

• BOK's Project Hangang (Phase 2 resumed March 2026) is Korea's CBDC programme — wholesale phase currently; no direct merchant acceptance path exists for foreign operators
• Digital Asset Basic Act covering crypto issuance and trading is expected in 2025–2026; KRW stablecoin activity from KakaoBank and Naver is the most operationally relevant development to monitor
• MyData expansion (personal data portability under the Credit Information Use and Protection Act) continues to extend consumer data rights — relevant for operators building data-intensive financial products in Korea
• The KFTC cross-border QR rollout (India, Vietnam planned after Indonesia live) may gradually expand the utility of Korean wallet acceptance for cross-border travel and e-commerce flows

Korea rewards operators who understand the local-PG structure and commit to it. The regulatory and fee environment is transparent and stable — the FSC publishes MDR reform decisions explicitly, and the KFTC's open-banking infrastructure is mature. The complexity is structural rather than opaque.