There is a useful heuristic for distinguishing genuine financial infrastructure shifts from speculative cycles: watch what the custody and clearing business does. The firms that hold assets, settle trades, and manage counterparty risk in the traditional financial system do not move toward new infrastructure until they believe it will persist. They are slow, conservative, and deeply incentivised to protect the systems that currently make them money. When they move anyway, it means the calculus has changed.

BlackRock moved. In March 2024, the world's largest asset manager launched BUIDL — the BlackRock USD Institutional Digital Liquidity Fund — on the Ethereum blockchain. The fund holds short-term US Treasury bills, cash, and repurchase agreements. It issues tokens representing shares in that fund directly on-chain. By early 2025, it had accumulated over $500 million in assets. Franklin Templeton's FOBXX, which launched earlier and operates on the Stellar and Polygon blockchains, passed $500 million on its own. Together, two of the most cautious names in institutional asset management have put over $1 billion of real financial assets on public blockchains.

This is not a proof of concept. A proof of concept is what you run before you commit. BlackRock and Franklin Templeton have committed. The question worth examining now is what they have actually built, why they built it this way, and what the trajectory of institutional tokenization looks like from here.

What Tokenized RWAs Actually Are

A tokenized real-world asset is exactly what it sounds like: a token on a blockchain that represents a claim on something that exists outside the blockchain. That something can be a US Treasury bill, a share in a money market fund, a parcel of real estate, a corporate bond, a barrel of oil sitting in a warehouse, or a piece of receivables from a supply chain. The token is not the asset. It is a digital representation of a legal claim on the asset, enforced by a combination of smart contract code and the legal structures wrapped around it.

The trust problem this solves is settlement. Traditional financial assets settle through a chain of intermediaries — brokers, custodians, clearinghouses, transfer agents — each of which is essentially a ledger keeper, maintaining records of who owns what and ensuring that when assets change hands, both sides of the transaction are fulfilled simultaneously. This chain is expensive, slow by technological standards (equities in the US settle T+1, meaning the day after a trade; many bond markets still settle T+2), and creates counterparty risk at every link.

Blockchain settlement is atomic — the transfer of the token and the transfer of the payment happen in a single transaction, or neither happens. There is no settlement lag, no counterparty exposure between trade and settlement, and no need for a clearinghouse to stand between buyer and seller guaranteeing performance. The trust is in the code and the chain, not in the institutions layered between the parties.

The traditional financial system does not have a technology problem. It has an intermediation cost problem dressed up as infrastructure. Tokenization exposes the underlying structure directly.

Why BlackRock Built BUIDL the Way It Did

BUIDL is structured as a traditional fund — a Section 3(c)(7) fund under the Investment Company Act of 1940, available only to qualified investors. It holds the same assets a conventional money market fund holds. The innovation is entirely in how ownership is recorded and transferred: on Ethereum, as ERC-20 tokens, with Securitize acting as the transfer agent managing the token issuance and redemption process.

The choice of Ethereum is deliberate and worth noting. BlackRock did not build on a private, permissioned blockchain. It did not build on a bank-controlled consortium chain. It built on the same public Ethereum network that hosts decentralised finance protocols, NFT marketplaces, and every other application in the Ethereum ecosystem. The reason is composability — the ability of the BUIDL token to interact with other protocols, contracts, and financial infrastructure that also runs on Ethereum.

A BUIDL token can, in principle, be used as collateral in a lending protocol. It can be transferred between wallets without going through a broker. It can be used as a settlement asset in an on-chain derivatives contract. The fund's yield — the interest earned on the Treasury bills it holds — accrues daily and is distributed in new tokens. None of this required building new infrastructure. It required putting a traditional asset on infrastructure that already existed.

That is the architectural insight institutional players have arrived at, somewhat later than the industry expected but with considerably more capital. The blockchain infrastructure does not need to be designed for institutions. The institutions need to learn how to use the infrastructure that already exists.

Franklin Templeton's Different Bet

Franklin Templeton's approach with FOBXX is similar in objective but different in execution. Where BUIDL launched on Ethereum and has since expanded to other chains, FOBXX launched on Stellar and subsequently added Polygon. Stellar is a blockchain built specifically for payments and asset issuance, with lower transaction costs and faster finality than Ethereum's mainnet. The choice reflects a different priority: FOBXX is optimised for transfer efficiency and cost, while BUIDL is optimised for composability with the broader Ethereum ecosystem.

Franklin Templeton has also gone further in making FOBXX transferable peer-to-peer without going through the fund's transfer agent — a meaningful structural difference from BUIDL, which currently requires Securitize involvement in transfers. The direction of travel is toward greater tokenisation, not less.

Both funds ultimately represent the same underlying thesis: US Treasury exposure, the safest liquid asset in the world, delivered with the efficiency properties of blockchain settlement. The market they are competing for is institutional cash management — the hundreds of billions of dollars that large organisations, funds, and protocol treasuries hold in short-term liquid assets and need to move efficiently.

What Comes After Money Markets

Money market funds are the easiest category of real-world asset to tokenize. The underlying assets are liquid, standardised, and already heavily intermediated. The legal structures are well-understood. The counterparties are sophisticated. The regulatory framework, while requiring navigation, is not hostile.

The more interesting question is what the next category is. The candidates with the most active institutional development are private credit, corporate bonds, and real estate.

Private credit — loans made directly to companies outside the public bond markets — is a $1.7 trillion asset class globally that is almost entirely illiquid. Investors who put money into private credit funds typically cannot exit for years. Tokenization creates the possibility of a secondary market, allowing investors to trade their positions to other qualified buyers without waiting for the fund's natural lifecycle. Several firms, including Hamilton Lane and KKR, have created tokenised feeder funds for private credit, though secondary market liquidity remains thin.

Corporate bonds are a closer-term target. The bond market is large, liquid at the top end but illiquid in the long tail of issuances, and operationally expensive — particularly for smaller issuances where the fixed costs of underwriting, custody, and administration consume a significant portion of the economics. A tokenised corporate bond issued directly on-chain, with coupon payments and principal repayment handled by smart contract, eliminates a substantial portion of that cost structure.

Real estate tokenization is further out, not because the technology is more difficult but because the legal structures are more complex. Property ownership is local, governed by state and national law, and typically requires physical title registration systems that have not yet been connected to blockchain infrastructure in any major jurisdiction. The economic case is clear; the implementation path is not.

The Structural Implication

The custodians, clearinghouses, and transfer agents who currently sit between asset issuers and investors are not unaware of what tokenization implies for their businesses. Some are building on-chain infrastructure themselves — BNY Mellon, State Street, and JPMorgan's Onyx division have all announced or deployed tokenisation-related products. Others are acquiring the firms that have built the relevant technology.

The pattern is familiar from other infrastructure transitions: incumbents first dismiss the new approach, then experiment with it, then acquire or replicate it, then advocate for regulation that preserves their position within it. We are somewhere between the experiment and acquisition phases for institutional tokenization. The regulation that follows will determine how much of the disintermediation the new infrastructure actually delivers, and how much gets recaptured by the incumbents through licensing, custody requirements, and compliance overhead.

BlackRock running a fund on Ethereum does not mean Ethereum has won. It means the conversation about which financial infrastructure persists has moved from whether blockchain has institutional utility to who controls the interface between legacy finance and on-chain settlement. That is a narrower question, but it is the one that will determine how much of the value capture stays with the new infrastructure and how much flows back to the institutions building bridges to it.