Author: Yi He

I. First, let's correct a misjudgment: This is not "another round of concepts," but rather a relocation of cash pipelines

When many managers see "$27.5 billion" and "30% growth in a single quarter," their first reaction is still: the scale is too small, far from a trillion dollars, let's wait and see. This judgment is mathematically correct, but strategically dangerous.

The key to RWA tokenization isn't how many billions of dollars are on-chain today, but rather who has moved the core processes of settlement, registration, issuance, and custody onto the programmable ledger. According to RWA.xyz, excluding stablecoins, on-chain RWA increased from approximately $21 billion at the beginning of 2026 to approximately $27.5 billion by the end of the first quarter, a quarterly growth of about 30%; among which, tokenized US Treasury bonds exceeded $13 billion in early April. By July, on-chain RWA, according to different statistical methods, was between $30 billion and $33.5 billion. The controversy lies in the method, not the direction: it's still small, but the growth rate is no longer like an experimental project, but more like the early stages of infrastructure migration. More importantly, it's about the structure. Early on-chain assets were driven by crypto-native institutions; now, banks, asset management companies, exchanges, custodians, and clearinghouses are entering the fray. The NYSE and Nasdaq are both designing trading capabilities for tokenized securities, SEC staff clarified in early 2026 that "tokenized securities are still securities," the EU's MiCA transition period has ended, and Hong Kong, Singapore, and the UAE continue to issue licenses. The question has shifted from "can it be done?" to "can your institution handle it?" Two or three institutions are doing three different things, not the same kind of "on-chain" approach. Putting JPMorgan Chase, HSBC, and Franklin Templeton in the same title might be misleading, suggesting they are all issuing tokens. On the contrary, they are respectively controlling three critical hurdles in the industrialization of RWA. JPMorgan Chase's Kinexys card focuses on "money." Its emphasis is not on retail tokens, but on institutional-grade deposit tokens, programmable payments, and cross-border settlements. Public information shows that Kinexys has processed over $4 trillion in transactions since its inception, with an average daily transaction volume exceeding $7 billion. JPM Coin-like deposit tokens are beginning to extend into public blockchain environments, serving institutional clients in moving funds between different currencies and time zones. For finance departments, the significance is direct: if cash can be settled 24/7, friction caused by weekends, holidays, and correspondent bank deadlines becomes programmable rules, rather than uncontrollable risks. HSBC's Orion card focuses on "vouchers." In February 2026, the UK Treasury selected HSBC Orion as the pilot platform for its Digital Gold Edge Bond (DIGIT) platform. HSBC disclosed that Orion has supported over $3.5 billion in digital-native bond issuance globally, covering sovereign, supra-sovereign, central bank, financial institution, and corporate issuers. Sovereign debt forms the top of the collateral pyramid in the financial system; when governments begin testing the placement of bond issuance, registration, and settlement on DLT, the signal is not "a bank has innovated," but rather that the underlying format of the bond market is being renegotiated. Franklin Templeton Benji is holding "fund shares." FOBXX/BENJI is one of the earliest US-registered money market funds to use a public blockchain as its official share registration system. By April 2026, the BENJI series of products had reached nearly $2 billion in size, with the number of investors growing by over 140% in two years, supporting peer-to-peer transfers, 365-day on-chain dividends, and second-level interest calculation. This proves one point: tokenization is not simply packaging assets as a crypto gimmick, but rather changing the way fund shares are registered, transferred, and their returns distributed.

III. The real debate is not "whether it will reach trillions," but who will first seize the fee-charging opportunity

Long-term forecasts differ greatly: McKinsey's baseline scenario for 2024 estimates tokenized financial assets at approximately $2 trillion by 2030, with an optimistic scenario of approximately $4 trillion; Citi's 2026 report places tokenized securities around $5.5 trillion; BCG and Ripple give even higher forecasts, at $9.4 trillion by 2030.

A discerning reader shouldn't get bogged down in which number is most accurate, but rather focus on the underlying judgment: the potential is tens of times greater than the hundreds of billions of dollars available today, compared to any credible 2030 target. The difference lies only in speed, not direction. Therefore, in the next three to five years, the focus should not be on "buying an RWA concept," but rather on four types of investable, constructible, and acquireable capabilities. First, tokenized cash management. This is the closest to being a "established" demand. For token holders, trading platforms, market makers, and multinational corporations, converting idle stablecoins or fiat currency into tokenized government bonds/money market funds essentially transforms unyielding or inefficient balances into settleable, collateralizable, and programmable cash equivalents. Its competitive advantage lies not in a few basis points higher yields, but in compliance access, redemption speed, custody arrangements, and inclusion in mainstream collateral frameworks. Second, private lending and securitization. This is the area most easily underestimated in terms of scale. The value of on-chain private lending, mortgage lines, trade finance, and asset-backed securities lies not in allowing retail investors to "buy some high-end assets," but in achieving standardization in lending, services, cash flow collection, default disclosure, and secondary transfers. Investment opportunities here come with higher due diligence requirements: the quality of underlying borrowers, service provider incentives, and data authenticity are all more important than the "token" itself. Third, infrastructure "tollbooths." When assets, cash, and compliance rules are all on-chain, the real money-makers are those providing custody, transfer agents, compliance whitelists, oracles, cross-chain interoperability, audit reports, fiat currency deposits and withdrawals, and institutional wallets. The investment logic is similar to the internet era: not betting on a single website, but holding cloud, payment, identity authentication, and security. The more RWA becomes mainstream, the more it needs this seemingly mundane but crucial middleware layer. Fourth, selective physical assets. Gold and commodities are ahead because their collateral logic is clear, pricing is transparent, and custody is mature; real estate, private equity, and art will lag behind because valuation, legal enforcement, and secondary liquidity are more difficult. Be especially wary of sales pitches like "tokenization equals liquidity" for real estate: the ability to transfer tokens in real time doesn't mean someone will be willing to take your commercial building shares at midnight on a Sunday.

IV. Risks are not on-chain, but off-chain still require human signatures

The biggest misconception about RWA is that on-chain transactions automatically eliminate counterparty risk. It cannot. It eliminates some friction in reconciliation, settlement, and transfer; it retains and even amplifies three other types of risks.

First, legal enforcement risk. Whether a token is a security, fund, deposit, bill, or commodity determines bankruptcy remoteness, holder rights, and cross-border sales rules. Second, operational risk. Private key management, smart contract vulnerabilities, oracle errors, and cross-chain bridge attacks can all turn "real-time settlement" into "real-time loss." Thirdly, there's the illusion of liquidity. Many tokenized assets still follow a minting-holding-redemption model, with inactive secondary trading. If your stress test assumes it can be sold at any time like a money market fund, you're misinterpreting infrastructure upgrades as market depth. For Chinese institutions, three additional things need to be distinguished: research and allocation of compliant overseas products, digital transformation of the income rights of domestic real assets, and unregulated tokens issued in the name of RWA. The first two can be tracked, but the third should be avoided. RWA's legitimacy comes from assets, licenses, custody, and disclosure, not from a beautifully written white paper. Fifthly, a 90-day checklist for finance and investment managers. Don't just hand over RWA to the innovation department to write a trend report and call it a day. This should be treated as an issue of working capital efficiency: First, assess which currencies, time zones, and banks within the group have cash holdings and settlement delays; second, conduct a small-scale real-money redemption test on a regulated tokenized money market fund or deposit token, focusing on observing the arrival path outside of business hours; third, update custody and counterparty policies, clarifying private keys, wallets, whitelists, bankruptcy remoteness, and audit evidence; fourth, ask each supplier three questions—who is the legal owner of the assets, where are the underlying assets located, and who is fulfilling the obligations on the other end when redeeming at 2 AM on Sunday. To put it bluntly: if your competitors can reduce cash transfers from T+1 to minutes, make collateral available 24/7 from frozen, and make fund subscriptions and redemptions a continuous market from business hours, then "we are still researching and piloting" is no longer prudent, but rather surrendering costs, speed, and customer experience to others. **RWA won't put all assets on-chain tomorrow, but it will rewrite the answer to the question of "how assets are held, settled, and pledged."** In investing, don't chase concepts, buy cash flow; don't be misled by liquidity, verify redemptions; don't ask how cool the blockchain is, ask who bears legal responsibility off-chain.