Blockchain for Korean Enterprises: Choosing the Right Chain and the Right Way to Build

Key Takeaways

• Chain selection should not start with a technical stack, but with five sequential decisions: whether blockchain is needed, whether the trust model should be public or private, whether to build or adopt infrastructure, which stack to use, and how regulation applies.
• The rationale for using blockchain has evolved beyond “mutual distrust” between parties; today, the stronger reasons are 24/7 programmable settlement, atomic delivery-versus-payment, seamless backend UX across asset types, and auditability that legacy systems cannot provide.
• Public chains no longer necessarily mean giving up issuer control, as institutions now commonly deploy assets on public infrastructure while retaining whitelist, freeze, reissuance, and transfer-control rights at the smart contract level.
• Global financial institutions are increasingly using public chains or institutional shared networks such as Canton, while fintech and payment companies are taking more open, multi-chain, or application-specific chain strategies.
• In Korea, chain strategy will be shaped heavily by regulation: stablecoins are likely to converge around bank-led issuance, token securities around KSD-linked infrastructure, and fintech players around distribution, application layers, or selective self-built chains.

This article is adapted from "Korean Blockchain Guidebook for Institutions 2026," jointly published by Four Pillars and Pantera Capital. The full report covers 14 more themes for companies and institutional investors.

1. Overview: The Chain Decision Comes Down to Five Sequential Questions

Enterprise blockchain adoption accelerated sharply through 2025. Stripe launched the mainnet of Tempo, a payment-specialized Layer 1 chain, in March 2026. JPMorgan formally migrated its deposit token JPMD to Base in November 2025, and in January 2026 announced native issuance on the Canton Network. The tokenized MMF from Goldman Sachs and BNY launched on Canton in July 2025, and Robinhood activated its own Layer 2 public testnet in early 2026, bringing its tokenized equity count to nearly 2,000. As fintech companies, financial institutions, and payment firms all move in earnest toward blockchain adoption, the questions practitioners face have grown more concrete and more complex.

The most common mistake practitioners make when planning a blockchain solution is to start from the choice of technical stack. Chain-level decisions actually come down to five sequential questions:

• Is a blockchain even necessary?
• Public or private trust model?
• In-house build or external adoption?
• Which specific stack?
• What does the regulatory environment allow?

The first question, whether a blockchain is necessary, has to be rigorously validated. The criteria for this validation have shifted substantially over the past five years. Mutual distrust among multiple organizations was once treated as the core requirement for blockchain, but that criterion alone has proven inadequate as a decision rule.

First, the premise of mutual distrust collapses easily. The moment a consortium forms, a governance structure is created, and centralized trust points re-emerge within it. Second, as blockchain infrastructure itself has matured, several new categories of economic utility have taken concrete shape:

• 24/7 programmable payment requirements: cases where the time constraints of traditional payment infrastructure, such as T+2, business-day limits, and time-zone-based cutoffs, become a bottleneck for business value.
• Automation of securities delivery and settlement: cases where there is a clear economic incentive to have securities delivery and settlement resolved simultaneously in a single transaction, without passing through a chain of intermediary institutions.
• Seamless backend UX: cases where UX improvements depend on handling in a single backend all the asset classes and activities that have until now been fragmented, including deposits, stocks, bonds, loans, points, and digital assets.
• Auditability unattainable with existing infrastructure: cases where onchain transparency itself lowers regulatory compliance costs or reduces trust costs.

The criteria for choosing between public and private chains have also shifted between the traditional view and the current one. In the traditional framing, this choice was defined by whether a common ledger had to be operated on the premise of mutual distrust among competing institutions. The logic was that high mutual distrust called for a public structure with decentralized consensus, and low mutual distrust called for a private structure with lower consensus cost.

In the contemporary framing, the same choice has been redefined as a tradeoff between public-chain liquidity and network accessibility on one side, and issuer control and privacy on the other. Most institutional token products today are deployed on public chains, but they are designed so that the issuer fully retains whitelist, freeze, and reissuance authority.

As a result, the situations in which private chains are still the right choice have narrowed further. They apply where the main use case already sits within the institution's own customer network and external liquidity access is not genuinely required, or in certain jurisdictions where public chains are regulatorily off-limits. Private chains are no longer the choice for trust-deficient environments; they are the choice for environments with specific control requirements or regulatory constraints.

The rest of this section examines how these two shifts are actually playing out across global blockchain deployment cases.

2. Global Trends: Blockchain Adoption Across Institutions, and the Patterns That Are Emerging

2.1 North American Financial Institutions

Among major global financial institutions, blockchain adoption shifted quickly from self-hosted private infrastructure to public-chain-based product deployment beginning in 2025.

• JPMorgan: uses public and private blockchains selectively, depending on asset class and transaction type. Since 2019, the bank has operated the private chain Kinexys (formerly Onyx), which has processed over $1.5 trillion cumulatively and more than $2 billion in institutional transactions on average per day. In June 2025, it pilot-deployed its USD deposit token JPMD on Base, the Ethereum L2 developed by Coinbase, and fully migrated in November. Then in January 2026, working with Digital Asset, it announced native issuance of JPMD on the Canton Network. JPMD is deployed on a public chain, but it is transferable only between whitelisted addresses designated by JPMorgan. Smart contract control, key management, role separation, and asset movement authority between all addresses sit exclusively with the bank. By contrast, in November 2025, JPMorgan completed the first tokenization of a private equity fund in collaboration with Citco, and that transaction was executed on the existing private chain Kinexys.
• Goldman Sachs, BNY Mellon, DTCC: these are the players most actively using the Canton Network. In July 2025, the first U.S. tokenized MMF solution, integrating Goldman Sachs' GS DAP and BNY's LiquidityDirect, launched on Canton. In December 2025, DTCC selected Canton as its official tokenization partner and is currently running a U.S. Treasury tokenization pilot on the network.
• BlackRock: rather than build tokenization infrastructure in-house, BlackRock chose a division of labor between asset issuer and infrastructure provider. It has entrusted the entirety of BUIDL's tokenization, compliance, and transfer agency functions to Securitize, allowing it to focus on its core asset management business.

2.2 European and Asian Financial Institutions

• Deutsche Bank: to secure both the verifiability of public infrastructure and the control of a private chain at the same time, Deutsche Bank turned to zero-knowledge proof technology. In November 2024, under Singapore MAS's Project Guardian, the bank announced that it would build its own Layer 2 blockchain based on ZK Stack Prividium.
• Societe Generale (SG-FORGE): adopted a strategy of using different chains for different asset classes. It has deployed USDCV across multiple public chains, including Ethereum, Solana, XRP, and Stellar, and in November 2025 issued a SOFR-linked short-term digital bond on Canton, completing the first U.S. digital bond issuance.
• DBS: a commercial bank that has built up its own blockchain capabilities step by step, expanding from private infrastructure out to public chains. DBS built the world's first bank-operated digital exchange, DDEx (DBS Digital Exchange), in-house in 2019. In October 2024, it launched DBS Token Services on an EVM-compatible private blockchain, providing real-time settlement and programmable payments to institutional clients. In August 2025, it extended tokenization from its private blockchain out to Ethereum, and began distributing structured notes through third-party digital platforms such as ADDX, DigiFT, and HydraX.

2.3 Fintech and Payment Companies

While financial institutions took a cautious, phased approach, fintech and payment companies in 2025 adopted blockchain infrastructure far more aggressively. Unlike financial institutions, they showed a clear tendency to use multiple public chains at the same time rather than tie themselves to a single network, and when they did build their own chains, they tended to design them as open networks.

• Stripe: in 2024, Stripe acquired stablecoin infrastructure company Bridge for $1.1 billion. In September 2025, the company jointly announced a payment-specialized Layer 1 blockchain called Tempo with the crypto VC Paradigm, and launched the mainnet in March 2026. Tempo is an EVM-compatible chain on which fees can be paid in USD stablecoins without a gas token, and it reserves payment-dedicated blockspace at the protocol level.
• Circle: in August 2025, Circle officially announced Arc, an L1 for USDC payments and capital markets, and launched a public testnet in October. Arc is an EVM-compatible chain that uses USDC as its native gas token, with features including sub-second finality, a built-in FX engine, and optional privacy features.
• Robinhood: in June 2025, Robinhood announced that it was building its own Layer 2 chain on the Arbitrum Orbit stack, and activated a public testnet in early 2026. It currently issues tokens for around 200 U.S. stocks and ETFs on Arbitrum One for EU customers, and as of December 2025, the total number of tokenized stocks was approaching 2,000. Its own chain is being designed as an RWA-optimized structure, natively supporting 24/7 trading, self-custody, and cross-chain bridging.
• Visa: rather than build its own chain, Visa is positioning itself as a multi-chain settlement hub. It currently supports settlement for USDC, EURC, USDG, and PYUSD across four chains: Ethereum, Solana, Avalanche, and Stellar. In December 2025, it launched a pilot program with Cross River Bank and Lead Bank for settling USDC within the U.S. on Solana. Visa is a design partner for Arc and participates as an anchor validator for Tempo, taking a network-neutral stance.
• PayPal: chose an approach that relies on distributed deployment across public chains, without building its own. PYUSD, which launched on Ethereum in August 2023 and is issued by Paxos, expanded to native deployment on Solana in 2024, on Arbitrum in July 2025, and subsequently on Stellar, and it has been indirectly extended to Tron, Avalanche, and Sei through the omnichain version PYUSD0, via LayerZero.
• Revolut: without building its own chain, Revolut focused on building a user-facing layer on top of multiple public chains. It supports stablecoins such as USDC and USDT across six chains (Ethereum, Solana, Tron, Polygon, Arbitrum, and Optimism), and shortly after obtaining a MiCA license in October 2025, launched a 1:1 fee-free exchange function between USD and USDC/USDT. Revolut's own GBP stablecoin is currently in pilot through the UK FCA sandbox.

2.4 Convergence Patterns at the Stack and Protocol Level

Taken together, these cases reveal a clear set of patterns.

• The separation of issuer control from the chain layer has become standardized. Public chains used to be synonymous with giving up control, but once models such as JPMD and BUIDL began deploying on public chains while retaining full authority separation at the smart contract level, that perception stopped holding.
• Financial institutions and fintech or payment companies show distinctly different trajectories. Financial institutions have adopted as standard the practice of deploying private tokens that retain issuer control onto public chains. Fintech and payment companies take a more open approach. The difference stems from how much network accessibility each business model actually requires. Unlike financial institutions, where closed transactions among institutional clients are central, fintechs have to connect large numbers of payers and payees within an open economy, so deploying across multiple chains and building an open ecosystem is inherently more advantageous.
• The early self-hosted private consortium network model is being substantially absorbed by new-generation institutional public L1s such as the Canton Network. Canton has privacy embedded at the protocol level, and numerous institutions participate as chain validators. It has effectively become the institutional public standard, processing roughly $4 trillion in annualized tokenized volume.
• Building infrastructure in-house is no longer a necessary requirement. BlackRock's decision to entrust all of BUIDL's tokenization to Securitize shows how vertical division of labor between asset issuers and infrastructure providers can work in practice. Not every player needs to directly build its own chain, compliance stack, and custody infrastructure, and the choice between building and outsourcing should track each player's core competencies.

2.5 Practical Criteria for Stack Selection

Drawing from these global cases, the practical criteria for stack selection can be summarized as follows. Once the direction has been set toward building one's own chain, the next decision is not which specific framework to compare. It is whether to be incorporated into the Ethereum ecosystem as a lower layer (L2 rollup), build an independent Layer 1, build an application on an institutional public network such as the Canton Network, or build a private chain.

2.5.1 Path 1: Building an Ethereum Layer 2 Chain

An Ethereum Layer 2 chain is a structure that inherits the security model of Ethereum mainnet while operating its own chain space. Ethereum mainnet acts as the settlement layer, and the rollup independently composes only the execution layer. The Ethereum Layer 2 framework landscape is a four-way contest among Optimism's OP Stack, Arbitrum's Orbit, ZKsync's ZK Stack, and Polygon's CDK.

• OP Stack: has the widest adoption base. It accounts for over 50% of transaction volume across the Layer 2 chain ecosystem, with more than 32 mainnet chains running on the stack. The developer pool is the largest, and the tooling is the most mature. Optimism, the developer of the OP Stack, launched its enterprise management solution OP Enterprise in January. In April, it added Privacy Boost, an enterprise privacy solution, as part of the OP Stack. Coinbase's Base is the best-known deployment built on OP Stack. Base alone generated roughly $250M in sequencer revenue in 2025.
• ZK Stack: a zero-knowledge proof-based framework. Its main advantage is that transaction integrity is proven through zero-knowledge proofs, which enables faster finality. ZKsync, the developer of ZK Stack, also provides Prividium, a privacy-preserving institutional variant of the stack. Deutsche Bank's asset tokenization platform DAMA 2 is currently being built on Prividium.
• Arbitrum Orbit: the defining feature is that it can be used to build a Layer 2 that settles directly to Ethereum, or a chain in the form of a Layer 3 on top of an existing Layer 2. It allows per-chain design of custom gas tokens, governance, and permissioned or public structures, and offers high technical flexibility, including support for Rust, C, and C++ based smart contracts beyond Solidity through Stylus. One notable use case is Robinhood, which built its tokenized stock platform as an Orbit-based Layer 2 and activated a public testnet in early 2026.
• Polygon CDK: a zero-knowledge proof-based framework whose defining feature is that the data availability layer, gas token, and execution environment can be freely configured on a per-chain basis. In 2026, CDK Enterprise was released with enhanced privacy and compliance features for financial institutions, and in March, Apex Group announced it would build a CDK-based chain called T-REX.

Whichever framework is chosen, operating a Layer 2 chain requires running a data availability (DA) solution in addition to the chain's own node client, and together these two account for a significant portion of L2 operational costs. Submitting data directly to Ethereum is the safest option but relatively expensive, and it tends to be the default choice when the purpose is institutional, high-value asset issuance. Borrowing external DA solutions such as EigenDA, Celestia, or Avail brings costs down, but each has a different security model, so proper evaluation is essential.

2.5.2 Path 2: Building an Independent Layer 1 Chain

An independent Layer 1 chain operates its own validator set, separated from Ethereum. Security rests on its own validator group, and all decisions regarding transaction finality, fees, and upgrades are vested in the chain itself. Independent Layer 1 is usually chosen in situations where an application's characteristics make fee volatility or transaction processing speed on existing chains a real constraint, or where custom gas tokens, permission models, or fee policies are required.

The commonly used stacks for building an independent Layer 1 are Avalanche L1 (formerly Subnet) and the Cosmos SDK. Avalanche L1 is an EVM-compatible L1 structure that operates an independent validator set and supports custom gas tokens and multi-level permission models. It grants a high level of control to institutions by letting them form their own validator group, so that only participating institutions verify transactions. Institutional references for Avalanche L1 are concentrated in proof-of-concept experiments, such as JPMorgan's Onyx Evergreen experiment and the Woori Bank-BDACS KRW stablecoin proof of concept. The Cosmos SDK is a meaningful alternative in the consumer application category, but institutional and enterprise references for it are currently limited.

A newer stack called Commonware has also attracted attention recently. Commonware observes that existing frameworks such as OP Stack confine chains to predetermined execution rules and security configurations, and it is built so that a chain can be assembled as a stack from which the necessary components are selected and combined. Commonware allows customization of almost every component of a blockchain, including consensus and signing mechanisms, state management databases, and fee calculation mechanisms. That makes it a particularly suitable stack for blockchain projects built with a specialized purpose, or for projects that require cryptographically flexible implementations. Stripe's payment-centric Layer 1 chain Tempo, mentioned above, is the most visible implementation to date.

2.5.3 Path 3: Joining an Institutional Public Network or Building a Private Network

The Canton Network is a Layer 1 chain with built-in privacy features, developed by Digital Asset. In the Canton ecosystem, each institution processes transactions in an isolated environment with transaction-level privacy embedded at the protocol level, and their consistency is verified by a small number of institutional validators (Super Validators). The practical effect is that transaction contents between competing institutions are not mutually disclosed.

Within the private network stack, what used to be a two-way contest between Hyperledger Fabric and Corda has now shifted, with Hyperledger Besu emerging as the main choice. Fabric is still a reasonable option for non-financial consortia and large-scale enterprise deployments, but in projects where tokenization and financial integration are central, Besu, which offers EVM compatibility, tends to be the primary pick.

3. Opportunities in the Korean Market: Regulatory Terrain and Execution Paths

Practitioners seeking to build blockchain solutions in Korea cannot apply the global decision frame described above directly. Korea's regulatory terrain restricts options at several branching points, and in some places forces specific paths. The first step is to map which statutes apply to each use case, and how they affect chain selection.

3.1 Regulatory Terrain

In the area of KRW stablecoins, the Bank of Korea formalized a bank-centric issuance principle in the 2025 Payment and Settlement Report, published in April 2026. The report describes a mutually complementary structure in which banks are responsible for issuance, while non-banks take on the roles of discovering innovative use cases, distribution, and user acquisition. The central scenario in the Digital Asset Basic Act currently being promoted by the government and the ruling party reflects this principle. It includes a 51% rule requiring banks to hold at least 50%+1 of the shares in the issuing entity, and a 100% reserve requirement, which together leave virtually no room for non-bank enterprises to issue stablecoins on their own.

In the area of tokenized securities, the regulatory path was substantially clarified with the passage of the revised Electronic Securities Act and Capital Markets Act in January 2026. The distributed ledger has been defined as a securities account book with legal effect, and a new "account-management institution for issuers" category has been established. Issuers that meet requirements such as KRW 1 billion or more in equity capital, rightsholder protection facilities, and conflict-of-interest prevention systems can now register with the Financial Services Commission and issue directly. Over-the-counter distribution of investment contract securities through securities firms has also been permitted. The Korea Securities Depository (KSD) has been operating a tokenized securities platform testbed since June 2025, which is equipped with distributed-ledger-based electronic registration eligibility review and issuance and distribution total-volume management functions. The expectation is that multiple account-management institutions will need to participate in the securities-dedicated mainnet, with linkage to KSD-led infrastructure set as mandatory.

The two areas share a common feature. The "absorption into an institutional public Layer 1" pattern observed globally appears in Korea as a centralization structure driven by public or semi-public infrastructure (a bank consortium or the KSD). Unlike the global market, where the Canton Network has become the institutional public standard largely through market forces, Korea limits issuance entities and infrastructure by law, so competition over chain selection itself is relatively restricted.

3.2 Domestic Trends

With issuance paths effectively narrowed to specific entities, the real competition in the Korean market shifts to the question of who takes on which role.

3.2.1 Financial Sector

Banks and securities firms are positioning themselves as issuance entities in stablecoins and as account-management institutions in tokenized securities. Each is building a different strategic position.

• KB Kookmin Bank: is most deeply involved in coordinating with the Bank of Korea-led public infrastructure. In April 2026, KB Financial Group signed a business agreement with the Bank of Korea for the "expansion of payment and settlement based on digital currency and deposit tokens." In Phase 2 of Project Hangang, it designed a system that, in partnership with KG Inicis, Korea's largest PG firm, enables deposit token payments on top of existing payment infrastructure.
• Shinhan Investment Securities: together with SK Securities and LS Securities, is aiming to become the standard platform for STO through the Pulse mainnet, developed by Blockchain Global.
• Woori Bank: is targeting the B2B and tokenized securities settlement layer. Its equity investment partner BDACS issued KRW1, a stablecoin with a 100% KRW-collateralized structure, on Avalanche L1 in September 2025. Woori Bank is emphasizing its use as a settlement means for inter-corporate payments and securities token transactions.

3.2.2 Fintech and Distribution-Touchpoint Players

Excluded from issuance, these players are instead trying to secure positions in adjacent layers such as distribution, infrastructure, and custody. Three representative paths emerge.

The first is securing the distribution and application layer by building one's own chain.

• GIWA: an Ethereum Layer 2 chain built on OP Stack, oriented toward a finance-friendly blockchain that bakes in KYC and AML considerations, and positioned as a hub for KRW and global stablecoin distribution. Dunamu is pursuing a strategy of occupying one axis of the stablecoin ecosystem as a distribution partner rather than as an issuer, leveraging its own trading infrastructure and liquidity. In February 2026, Dunamu completed a proof of concept with Hana Financial Group in which SWIFT-style remittance messages between Hana Bank's domestic and overseas branches were replaced by blockchain messages on the GIWA chain. In the course of that PoC, it verified a structure that protects user information by applying "BOJAGI," a zero-knowledge proof-based privacy protocol.
• Maroo: unlike GIWA, Maroo chose to build an independent Layer 1 that is not incorporated into the Ethereum ecosystem. It focuses on a range of KRW stablecoin use cases, including paying transaction fees in the KRW stablecoin, coexistence between regulatory-compliant paths and restriction-free wallet paths, and an AI-agent-dedicated account structure. Beyond stablecoins, it also targets an integrated platform for real-world asset tokenization and security token offerings. Some of its core technology is applied in "Bidanjumeoni(which means ‘Silk pouch’ in Korean)," a digital wallet launched in 2025 in collaboration with the Busan Digital Asset Exchange (BDAN).

The second path is securing a distribution network through alliances among big techs, or between big techs and exchanges. With the banking sector locked in as the issuance entity, big techs and exchanges are carving out entry paths through partnerships with issuers and through combined distribution networks.

• Naver-Dunamu alliance: Naver and Dunamu have partially completed acquisition procedures through a comprehensive stock exchange. The result is a stablecoin issuance and distribution structure that takes shape across Naver, Naver Financial, and Dunamu. Naver Pay's payment and commerce infrastructure, Upbit's exchange liquidity, and the GIWA chain are expected to combine into a single stack.
• Kakao Group: has formed a stablecoin task force spanning Kakao Pay and Kakao Bank, and is envisioning a business that covers issuance, distribution, payment, and custody using their infrastructure. Kakao Group accumulated relevant know-how in the past through the development and operation of Klaytn.
• Bithumb-Toss alliance: discussions began around a structure in which Bithumb would handle distribution and trading, and Toss the payment and remittance network. Progress has reportedly been slow because of IPO issues at both companies. Bithumb has partly pivoted toward expanding the distribution of dollar stablecoins to compensate for delays on the KRW side, and in April 2026 signed an MOU with Circle covering cooperation on digital asset infrastructure and stablecoin technology.

3.2.3 Execution Tasks: How Should Blockchain Be Introduced

3.2.3.1 Build a Chain or Deploy on an Existing One

The first gateway to blockchain adoption is deciding whether to build one's own chain or deploy on an existing chain. Put another way, the question is whether the users and traffic needed to sustain the chain already exist, and whether chain-level design control is directly tied to business competitiveness.

The first half of this question concerns economies of scale. Running one's own chain's sequencer or validators, building a developer ecosystem, integrating with bridges and wallets, and conducting security audits generate fixed costs on the order of several billion won per year. Offsetting those costs requires a user pool large enough that continuous transaction inflow makes the chain profitable from day one.

A common auxiliary path for offsetting fixed costs is token issuance. In public-chain cases, a portion of the initial construction cost is raised from external capital through token sales and incentive programs. Base is the notable exception, covering costs with sequencer revenue and its parent company's financial resources rather than a token. That model was only possible because Coinbase had already accumulated a large user base it could migrate to Base. Token issuance offers clear capital-raising advantages, but it is not a path that can be chosen on that benefit alone, and in Korea specifically, the execution risk of raising capital through this path is very high.

The second half of the question concerns the structural requirements of the product. Chain-level control becomes essential when a company wants to handle, on a single backend, asset classes that have until now operated in fragmented systems, such as deposits, stocks, bonds, loans, points, and digital assets. Redesigning the fee structure to fit one's own services, internalizing a model in which movement between asset classes is resolved in a single transaction, or reflecting Korean regulatory account structures and privacy at the protocol level is difficult to achieve on the default design of a general-purpose public chain. By contrast, if the main goal is simple token issuance or the distribution of a single asset, the same level of control can be secured through permission design at the smart contract level alone.

3.2.3.2 Fintech Platforms

In Korea, the domain that satisfies both conditions at once is large fintech platforms. Payments, remittances, points, and asset management already occur at large scale inside the platform, so initial demand for the chain is effectively guaranteed. Looking ahead, the need to handle stablecoin payments, tokenized securities trading, real-world asset tokenization, and AI-agent payments together on a single backend makes chain-level control practically necessary. Given the cost of chain operation, only players with a crypto-friendly user base such as Dunamu, or players with enormous general-purpose user pools such as Naver, Kakao, and Toss, can realistically consider launching their own chain. For fintech firms with smaller user bases, or whose asset-class integration needs stop at payments and points, deploying on existing chains is the realistic path.

Once the decision to build one's own chain has been made, the conversation shifts to the character of the stack. If the priority is to absorb the liquidity and developer pool of the Ethereum ecosystem, building on an Ethereum Layer 2 framework is the fast and safe option. If fees and privacy have to be redesigned at the protocol level, an independent Layer 1 is more appropriate.

For players who choose to deploy on existing chains, chain selection is driven by use case. If general-purpose payments and remittances are the focus, Ethereum, which offers the deepest stablecoin liquidity and the broadest wallet and exchange connectivity, is the default choice. Because Ethereum mainnet's gas costs are unsuitable for small-value payments, actual deployment typically happens on a Layer 2 that inherits Ethereum's security and liquidity while offering far lower fees. Outside of Ethereum, Solana, whose transaction speed and fee structure are particularly well-suited to small payments, is a realistic alternative. In practice, Visa's USDC settlement pilot, PayPal's PYUSD deployment, and Revolut's multi-chain support all include Solana. If the primary use case is cross-border remittance, payment-specialized Layer 1s such as Circle's Arc or Stripe's Tempo are worth considering.

Whichever path is chosen, it is wise to avoid a design that is tied to a single chain from the start. Placing an abstraction layer in the backend that assumes future expansion to multiple chains is the more prudent approach. Designing so that the user touchpoint and internal systems are not significantly disrupted when issuance chains are changed or added may be a more important decision than the chain selection itself.

3.2.3.3 Financial Institutions

The situation for financial institutions is different. Their core businesses of deposits, lending, and securities trading are mostly tied to off-chain systems, so even a large customer base does not automatically translate into the kind of on-chain demand needed to operate a dedicated chain. Most issued assets can be managed with permission control at the token level, so use cases that genuinely require chain-level control are few. Reflecting these conditions, the global standard has become to deploy tokens on existing public chains or on institutional public Layer 1s such as the Canton Network, retaining only issuer control at the smart contract level. The BlackRock-Securitize case also shows that vertical division of labor between asset issuers and infrastructure providers is a valid path.

For Korean financial institutions, applying the same two questions for the build-or-deploy decision would point to deploying on existing chains as the rational choice, but Korea's regulatory terrain prevents that choice from working directly. If the Digital Asset Basic Act draft passes in the current bank-centric scenario, the room for independent issuance by non-banks becomes very narrow. Several details remain unclear, including whether a structure in which a bank-led consortium issues tokens on a public chain is permitted, and whether distribution to domestic users and distribution to overseas users are regulated separately. On the tokenized securities side, the distributed ledger has been given legal effect as a securities account book, but the scope and depth of the linkage with KSD-led infrastructure, the question of chain-technology neutrality, and the permissibility of cooperation with external infrastructure providers are all still being specified.

Korean financial institutions therefore find themselves in a situation where on-chain demand is not yet large enough to justify running their own chain, regulation has not yet been set up to allow free deployment on existing public chains, and institutional public chains are also absent. The infrastructure for major asset classes is therefore likely to converge on a direction in which industry-level consortia jointly build permissioned public networks. By global standards, this is equivalent to building one's own chain. It differs in character, however, because it is not an individual-institution decision. It is a collective construction forced as a consequence of regulatory design.

Within these constraints, the actions that Korean banks and securities firms can take come down to four.

• The first is leading consortium construction. This means rallying other institutions within the industry and claiming the initial design and governance of the public network. Shinhan Investment Securities' move to organize the Pulse consortium together with SK Securities, LS Securities, and Blockchain Global, asserting standard-platform status even before tokenized securities legislation had passed, is an early example of this path.
• The second is positioning within the consortium. Industry-level public infrastructure divides roles among issuance entities, account-management institutions, settlement, custody, and distribution touchpoints. If the regulatory issues described above are confirmed along the central scenario, convergence on this path is expected to strengthen.
• The third is infrastructure outsourcing. This means entrusting technical infrastructure such as chain operation, tokenization engines, and compliance stacks to specialized providers, while the financial institution focuses on its core business of asset issuance, customer relationships, and credit and risk judgment.
• The fourth is an independent domain outside the consortium. Domains that fit here are ones that can be designed as the permissioned infrastructure of a single institution or a small number of partners, such as remittances between overseas branches, programmable payments for corporate clients, and loans collateralized by tokenized assets. In cross-border payments, because stablecoins sit in a regulatory vacuum under the Foreign Exchange Transactions Act, there is still room to discuss the use of public chains, but recent legislative discussion has been converging on incorporating them into the existing foreign exchange regulatory system.