Ask what "tokenization" actually means and you get a lot of hand-waving, so it helps to look at what is being tokenized at real scale versus what people talk about at conferences. The category is broad in theory, but the money clusters in a few buckets. US Treasury bills and money market funds are the biggest, led by BlackRock's BUIDL, Franklin Templeton's BENJI, and Ondo Finance's USDY. Private credit is second, with protocols like Centrifuge and Goldfinch lending on-chain to off-chain borrowers. Real estate exists but stays small, boxed in by regulatory complexity and the messiness of fractionating an actual building.
The numbers, and what they leave out
Tokenized RWAs went from under $1 billion in early 2023 to over $10 billion by early 2025, and that excludes stablecoins, which are really the original tokenized RWA. Impressive growth, but a rounding error against traditional finance. The US Treasury market alone is north of $25 trillion. So the open question is whether this stays a niche or grows into a real slice of TradFi assets.
Why anyone bothers tokenizing at all
The bull case comes down to a few efficiency wins over legacy plumbing. Settlement speed is the most obvious one. Traditional securities settle T+1 or T+2, so one or two business days. Tokenized assets settle in minutes on-chain. Faster settlement means less counterparty risk and capital that is not stuck in limbo waiting to clear.
Then there is fractional ownership. A lot of these assets normally come in chunky denominations, and some institutional products want $100,000 or more to get in the door. A tokenized version splits down to any size, so a $10 position in the same underlying becomes possible. That widens the buyer base and can sharpen price discovery.
Composability is the crypto-native part, and it is the one I find most interesting. Once a T-bill lives as an ERC-20 token, you can post it as collateral on Aave, swap it on Uniswap, or wire it into any DeFi protocol. Using a tokenized T-bill as collateral for an on-chain loan gets you Treasury yield and leverage at the same time, which is just not a thing you can build inside traditional settlement rails.
The legal reality does not go away
Wrapping an asset in a token does not change what the asset legally is. A tokenized Treasury bill is still a security under securities law. The token is a digital representation, but ownership, custody, and compliance all still happen off-chain, which keeps you leaning on the usual intermediaries: custodians, transfer agents, regulators. That is the ceiling on how "decentralized" any of this can honestly claim to be.
The issuer of a tokenized Treasury fund is a regulated entity. They run KYC on holders, restrict transfers to whitelisted addresses, and can freeze or seize tokens if the law says so. That is a different animal from a permissionless token like ETH or UNI, where anyone can hold and move without asking permission. The push and pull between compliance and DeFi composability is the whole design problem in one sentence.
So most RWA products today are pretty centralized. BUIDL can only sit in approved wallets. Redemptions run on the issuer's willingness and ability to process them. The chain gives you the settlement rail, but the trust model is the same as before: you trust the issuer, the custodian, and the regulator. What the chain adds is transparency, since every transfer is visible, and speed. It does not remove the need to trust someone in the middle.
Where RWAs bump into DeFi
The convergence is the part worth watching. MakerDAO has parked a big chunk of its reserves in tokenized Treasuries, earning yield on its backing instead of sitting on idle stablecoins. That makes DAI partly backed by US government debt, which is a strange and telling place for decentralized and traditional finance to meet.
Several lending protocols now take tokenized RWAs as collateral, which sets up a loop: the traditional asset earns yield on-chain while doubling as collateral for crypto-native borrowing. The underlying yield, say 5% from T-bills, effectively subsidizes the cost of borrowing on-chain and makes the economics look good for DeFi users.
The catch is that you are also importing TradFi failure modes into DeFi. If an RWA issuer blows up, cannot redeem, or freezes assets, every protocol holding those tokens as backing or collateral feels it. A big RWA token going bad could cascade through DeFi the way a stablecoin depeg does.
What this means if you actually trade
For crypto traders this cuts two ways. On the opportunity side, there are tokens levered to RWA adoption. Protocols that do the tokenizing, like Centrifuge, Ondo, and Maple Finance, and chains that attract the issuance, like Ethereum, Stellar, and Avalanche, can see more demand as the category grows.
The structural shift is that tokenized RWAs drag the traditional risk-free rate on-chain. Once you can earn 4 to 5% from a tokenized Treasury without leaving the chain, any DeFi yield below that stops making sense on a risk-adjusted basis. Why eat smart-contract risk for 3% stablecoin lending when a tokenized T-bill pays 5%? That puts a floor under DeFi yields and squeezes the premium DeFi has to offer over the risk-free rate to pull in capital. On Blockcircle we already see users comparing on-chain yields against that risk-free baseline, and once the baseline is sitting right there in the same wallet, the comparison gets a lot more ruthless.
The bigger-picture version is a large pool of traditional capital reaching crypto-native products. If institutions can hold tokenized Treasuries on-chain and then use them as collateral in DeFi, the addressable capital for DeFi jumps by orders of magnitude. Whether that actually lands depends on regulation, infrastructure maturing, and institutions getting comfortable settling on a blockchain, so for now I would watch the redemption terms and issuer disclosures on any RWA token before treating its yield as free money.