DTCC is now conducting a tokenization trial with about 40 companies, including JPMorgan, Goldman Sachs, BlackRock, Vanguard and the NYSE, to represent stocks and government bonds on-chain, tokens that only become useful collateral if a credit market can answer who is pricing them and what happens if the venues behind that price go quiet.
DefiLlama estimates the on-chain RWA market cap above $51 billion, and those same assets only generate nearly $3.8 billion in DeFi active total value locked (TVL), an occupancy rate of almost 7.7%.
The price becomes the gatekeeper
A credit market needs a feed, a set of locations from which the feed comes, and rules for what happens when those locations go quiet to price assets like tokenized stocks, bonds, and gold.
Someone must choose the oracle, test its independence, limit exposure, and decide when liquidations occur.
Katana Network CEO Matthew Fisher said an oracle’s configuration starts with the locations it pulls price data from at launch, and teams upgrade these as liquidity migrates to newer or deeper locations.
For newly listed tokens, that upgrade is lagging behind, because liquidity has not yet concentrated in one trusted location.
Fisher said institutions delegate that research to professional curators, the vault managers like Steakhouse and Gauntlet that evaluate collateral, approve markets and set exposure limits to Morpho, or to protocols like Aave that directly build their own oracle relationships.
He noted:
“The institutions appreciate that there is a professional atmosphere in the room.”
A December 2025 study on decentralized credit found that a small number of trustees managing ERC-4626 vaults now broker a disproportionate share of the total escrow value, concentrating underwriting decisions in that layer of the stack.
Fisher’s account of institutional behavior is consistent with what the data already shows independently.
He said a single oracle manipulation within a market that trusted a curator could tarnish that curator’s entire track record. A trustee who brings a damaged file to an investment committee will get what Fisher called “a hard no,” regardless of how it performs elsewhere.
Who pays if it fails?
Fisher described the trustee as the party that owns the risk decision and assumes the reputational and commercial consequences when a market breaks.
The depositor typically absorbs the financial loss directly, and pool-based models like Aave or isolated markets on Morpho often leave the underlying protocol without direct liability at all.
April’s KelpDAO exploit exposes this mismatch, with Aave’s board estimating $230 million in bad debt based on the related rsETH position, which originated outside Aave’s own codebase, with the Umbrella module absorbing approximately $50 million as the first line of defense.
This gap in accountability raises concerns about institutions that trust trustees, whose main punishment for a bad decision is reputation, while the saver eats the first dollar of his loss.
First-loss capital, mandatory insurance, recovery of fees, and disclosure of verifiable risks are the types of requirements that could shut this down.
Bitcoin is constantly traded in deep global locations, so the oracle design focuses on aggregation and resistance to manipulation. Tokenized stocks, bonds and commodities inherit a market calendar that still observes their reference assets.
Fisher said there is “no objectively correct approach” to pricing these assets once the primary market closes.
Some platforms calculate a moving average based on market maker quotes once trading stops. Binance has historically leaned on funding rates to influence weekend prices before announcing new plans for that approach this year.
Katana routes gold, silver and oil through Chainlink and closes those markets to new positions as soon as the underlying exchange closes. Traders can still reduce existing positions, and the isolated margin includes any subsequent losses.
The London Stock Exchange is planning an overnight session, LSE 24, for 2027, alongside Nasdaq’s move to 23-hour weekday trading and Cboe’s proposed 23×5 US stock trading. Weekends, trading halts and asset-specific gaps are excluded from all three plans.
The adoption test
In the bull scenario, platforms standardize after-hours pricing, circuit breakers, capital at first losses, and trustee disclosures over the next few years.
Citi expects tokenized assets to reach $8.2 trillion by 2030 under the bull scenario. If DeFi usage rises to 12% to 18% in that world, RWA-linked DeFi active TVL could land nearly $1 trillion to $1.5 trillion, turning tokenized Treasuries, equities and commodities into true collateral primitives.
In the bear case, tokenization continues to expand in terms of issuance, without resolving its governance layer.
Citi’s bear scenario estimates tokenized assets at $2.7 trillion by 2030. If DeFi usage stays within the 2% to 4% range that current data implies, RWA-linked DeFi active TVL lands closer to $54 billion to $108 billion.
Tokenized assets are piling up on balance sheets, and DeFi lending and composability have little impact on that.
Fisher noted that the institutional oracle’s sensitivity to tokenized stocks, bonds or commodities whose underlying markets close on weekends is greater than for crypto-native assets.
Institutions need a governance structure around their price feeds that is durable enough to survive an investment committee, along with a definitive answer to who will take the loss on the day a feed goes wrong.