Korea's bank-led stablecoin plan rewards banks for tokenizing the deposits they already hold

For a bank, a deposit is funding it can lend against and a customer it keeps ^[7][11]. A won stablecoin from an independent issuer competes for both ^[7][11]. A tokenized deposit puts the same money on a blockchain and leaves both on the bank's books ^[7]. "It's entirely rational for them to skew toward tokenized deposits," said John D'Agostino, head of strategy for Coinbase Institutional ^[8].

Anthony Koo, head of payments at Singapore-based StraitsX, spent much of his career in banking. At Wednesday's Seoul roundtable he said major banks already process payments efficiently through established networks ^[1][5]. "They already have an infrastructure on that side. So when they're presented with an alternative, they're trying to understand where the utility comes from," he said ^[6].

Licensing the issuer also picks the product. If licenses go only to tier-one banks, the likely product is the tokenized deposit, and payments stay on the existing networks ^[5][8]. "If you give licenses to tier-one banks, as most regulators are inclined to do, there's little incentive to push the boundaries in building out infrastructure compared with giving them to a fintech," Koo said ^[4]. John Cho, chief stablecoin officer at Kaia DLT Foundation, said incumbents fear losing market share and revenue ^[9]. He offered a compromise: banks provide custody and other firms develop services ^[10]. He stressed that the arrangements under discussion were not final ^[10].

Cross-border use is the third variable. Koo said regulators must balance capital controls and the won's restricted currency status with rules that work alongside those elsewhere in Asia ^[13]. "If you maintain the same capital controls you have with traditional legacy infrastructure, you're going to make all the optimizations and efficiencies you could see with blockchain infrastructure moot," Cho said ^[14]. Gunnar Jaerv, chief operating officer at Hong Kong-based First Digital, disputed the regulators' assumption. "Regulators very often think that if they open the floodgates and let stablecoins run freely, that would be a way to circumvent those capital controls," he said. "I don't think that's necessarily true." ^[15]

Any won token would enter a market dominated by dollar tokens ^[16]. In Cho's version, companies keep dollar assets for returns and use local-currency tokens for operating payments, provided settlement gets cheaper ^[17]. Jaerv pointed to Korea's manufacturing and investment ties with Vietnam, where money follows "the path of least resistance" ^[18]. Elsewhere in Asia, Koo said, international visitors already buy watches, handbags and collectibles with stablecoins so merchants can avoid higher card processing fees ^[19]. "There's demand for certain transaction types to be settled in stablecoins," he said ^[20].

Each of those uses needs settlement that costs less than the networks already in place ^[17][19]. I think the bank-led model gives that job to the firms with the least reason to do it, because an issuer that runs established payment networks gains little by undercutting them ^[2][5]. The executives did not present data showing Korean users moving to foreign or offshore tokens, so that part of the case against bank-only issuance is a forecast. I'd expect a smaller near-term effect: dollar tokens keep the cross-border business because the won product does not beat them on cost ^[16][17].

The opposing case is also about money. Ashley Moore, managing director at Hilbert Capital, said interest-paying stablecoins that drew money away from deposits could pressure banks' funding and ability to lend ^[11]. "I think we're worried about the unintended consequences, the second- and third-order effects," he said ^[12]. The Bank of Korea cites banks' compliance infrastructure and risk management ^[3]. Its model protects the deposit base that, by D'Agostino's account, pushes banks toward tokenized deposits in the first place ^[7].

This view is wrong if a bank-issued won token reaches merchants for less than card processing costs. It is also wrong if Korea adopts Cho's custody-and-services split and other firms build the payment layer on top of bank custody ^[10][19].