Bank of Korea Scales Up CBDC Pilot With Half a Million Users
The Bank of Korea will launch Phase 2 of its CBDC pilot in September, expanding from 81,000 wallets to up to 500,000 users across nine banks, with real government subsidies on the line.
Intelligence analysis by Llama
South Korea's central bank is scaling Project Hangang from a 2025 pilot that drew 81,000 wallets to a September Phase 2 with nine banks, biometric payments, and government disbursements for up to 500,000 users.
Imagine the government handing out special digital coins through your regular banking app. South Korea tried this with 81,000 people earlier this year, but most of them just held the coins and never spent them. Starting in September, the country is going to try again with up to 500,000 people and real government money, so it can see if people will actually use the digital coins like real cash.
Analysis
From a Three-Month Sandbox to a 500,000-User Live Test
Project Hangang's first phase ran from April through June 2025 and, by the Bank of Korea's own accounting, looked modest: 81,000 wallets opened and 114,880 transactions processed. The headline number that quietly undercuts the milestone is even smaller. Only 42% of those wallet-holders actually spent any of the deposit tokens they received, meaning a clear majority held the digital won as a curiosity rather than a working medium of exchange. The central bank's decision to nearly sextuple the user ceiling for Phase 2, while widening the bank count to nine, is therefore less a victory lap than an attempt to test what Phase 1 could not, namely whether ordinary Koreans will treat a central-bank liability as money once it sits inside a familiar banking app.
Adding the Pieces a Sandbox Cannot Mimic
The September expansion is structured less like another experiment and more like a dress rehearsal. According to the Bank of Korea, Phase 2 will introduce biometric payments, person-to-person transfers, and, crucially, real government subsidy disbursements routed through the token system. Each of those features addresses a specific gap exposed in the spring pilot. P2P transfers are the only way to test whether the token circulates outside a one-way government-to-citizen flow, which is where any real currency proves itself. Disbursing actual won-denominated subsidies forces the infrastructure to handle reconciliation, anti-fraud checks, and settlement timing that simulated transactions can paper over. Biometric payment rails, meanwhile, are an attempt to lower the friction that likely contributed to that 58% non-spending rate, since a CBDC that requires a separate app or passcode has little reason to displace a card tap.
The Deposit Token Architecture and the Stablecoin Question
Project Hangang is built on the deposit token model, meaning commercial banks issue the on-chain representation of central bank money rather than the central bank itself maintaining a direct ledger for end users. That architecture is doing strategic work. It positions Korean banks as the distribution layer for any future retail digital won, and it implicitly draws a line between state-backed tokenized deposits and privately issued stablecoins such as USDC or USDT, which face an entirely different regulatory regime in Seoul. For the broader crypto market, the relevant signal is not the 500,000-user figure but the design choice: a CBDC that funnels through existing banks narrows the use cases a stablecoin would otherwise occupy in domestic remittances, retail savings, and government payment rails. Phase 2 will not settle that contest, but it will generate the first large-scale behavioral data set from a G20 economy on how citizens actually treat a tokenized sovereign currency when it is presented as ordinary money.
Key points
- Phase 2 of Project Hangang launches in September with up to 500,000 users across nine banks, up from 81,000 wallets in the spring pilot.
- Phase 1 (April-June 2025) processed 114,880 transactions, but only 42% of wallet-holders actually spent any of their deposit tokens.
- Phase 2 introduces biometric payments, person-to-person transfers, and real government subsidy disbursements.
- The pilot uses a deposit-token model in which commercial banks, not the central bank, issue the on-chain representation of digital won.
- The design implicitly distinguishes state-backed tokenized deposits from privately issued stablecoins, a boundary that could shape Korea's crypto regulation.
If the 500,000-user test produces materially higher spending rates and smooth subsidy disbursements, the Bank of Korea would gain the empirical basis to push Project Hangang toward a broader retail launch, giving Korea a working state-issued digital payment rail that banks already understand. A successful design could also become a reference point for other Asian central banks weighing similar deposit-token architectures.
The 42% spending rate from Phase 1 is a warning that utility, not technology, is the bottleneck. If Phase 2's biometric and P2P additions fail to meaningfully lift active usage, the pilot risks producing evidence against retail CBDCs rather than for them, and could leave Korea's banks holding infrastructure investment with no clear path to scale.
