Why Is Only 10% of Tokenized RWA Used in DeFi?
Roughly $2.7 billion of about $27 billion in tokenized assets under management is actively deposited across DeFi lending markets — around a tenth. The shortfall is usually explained as immaturity, as though the remaining 90% is queued behind integrations nobody has built yet. The real cause is a design conflict. A lending protocol must be able to seize collateral from a defaulting borrower and sell it to whoever bids; a permissioned token restricts who may receive it. When the liquidator is not eligible, the transfer reverts and the liquidation fails, leaving the protocol holding an asset it cannot sell. This guide sets out where the composable capital actually is, how the protocols that do accept restricted assets solve the problem, and what closing the gap requires without breaking the compliance the restriction exists to provide.
TL;DR — Key Takeaways
- ✓The Gap: About $2.7 billion of roughly $27 billion tokenized is deposited in DeFi lending — near 10%. Some measures put it closer to $2.47 billion.
- ✓The Cause: Liquidation needs an unrestricted transfer to whoever bids. A permissioned token reverts that transfer. The compliance property breaks the recovery path.
- ✓How It Is Solved: The protocol becomes permissioned, not the asset unrestricted. Aave Horizon ran roughly $550 million in deposits as a controlled market by early 2026.
- ✓What Gets Used: The most standardised assets. BUIDL at about $2.9 billion across reserves, margin and lending; Morpho on Avalanche near $875 million of RWA exposure.
- ✓Half the Gap Is Correct: For a qualified-purchaser credit interest, unrestricted liquidation would break the exemption. That part of the gap is the rule working.
A Gap That Is Not Waiting on an Integration
About $2.7 billion of roughly $27 billion in tokenized assets is actively deposited across DeFi lending markets. The usual explanation is that the rest is early — that integrations are pending and the ratio will correct itself. That reading misdiagnoses the cause.
The obstacle is a direct conflict between two requirements. A lending protocol must be able to take collateral from a borrower who defaults and sell it to whoever will pay. A permissioned security token must be able to refuse a transfer to anyone who is not eligible to hold it. Both are correct on their own terms, and they cannot both hold in the same transaction.
Many tokenized assets carry transfer restrictions, KYC requirements or jurisdictional limitations “that make them difficult to plug into permissionless protocols” — compliance rails still limit open-market use.
— Analysis of RWA composability, 2026
The word doing the work is permissionless. A protocol that assumes any address can receive any asset has assumed away the defining property of a regulated instrument.
Where Exactly the Conflict Bites
Not at deposit. A permissioned token can usually be deposited into a protocol contract if that contract is whitelisted. The failure happens at liquidation, when the asset must move to a party nobody vetted in advance.
| Step | What the protocol needs | Does a permissioned token allow it? |
|---|---|---|
| Deposit collateral | Transfer from holder to protocol contract | Yes, if the contract is an eligible holder |
| Value the position | A reliable price for the collateral | Depends on the asset — NAV may be struck daily |
| Accrue and repay | No movement of the collateral | Yes — nothing transfers |
| Liquidate on default | Transfer to an arbitrary liquidator who bids | No — the transfer reverts if they are ineligible |
| Return collateral | Transfer back to the original holder | Yes — they were eligible to begin with |
Key Insight
Four of the five steps work. That is what makes this failure mode dangerous rather than merely limiting — a protocol can integrate a restricted asset, accept deposits, originate loans and run for months without touching the broken step. The defect surfaces on the first default, which is also the moment the protocol most needs the mechanism to function. An integration that has never been tested through a liquidation has not been tested, and the absence of incidents to date is a statement about credit conditions rather than about the design.
The Protocol Becomes Permissioned, Not the Asset Free
Every working RWA lending venue has resolved the conflict in the same direction: the protocol adopts the asset's restrictions rather than the asset abandoning them. Aave Horizon reached roughly $550 million in deposits by early 2026 operating as a controlled market rather than a general pool.
A defined liquidator set
The protocol maintains a list of vetted parties eligible to receive the collateral on a default. Liquidation still functions because every possible recipient was pre-approved, which is the only way an unrestricted seizure and a restricted asset can coexist.
Separate markets, not the general pool
Restricted collateral sits in a dedicated market with its own participants and parameters. This keeps the failure mode contained: a liquidation problem in an RWA market does not become a solvency question for an entire protocol.
Concentration in the most standardised assets
BUIDL, at roughly $2.9 billion, appears across reserves, margin and lending, including as margin posted by institutional traders on Deribit. Janus Henderson's JAAA and Apollo's ACRED also recur. Assets with clear valuation and a known holder base integrate; bespoke ones do not.
Growth where the rails already fit
Morpho on Avalanche reached around $875 million of RWA exposure, and the pattern is that deployment concentrates where an asset's restrictions and a venue's controls were designed to meet. This is composability by prior arrangement rather than by open access.
Composability by prior arrangement is a real capability and a smaller claim than the one usually made. It means a tokenized asset can be used as collateral at venues that have specifically prepared for it, which is closer to how conventional collateral works than to the open composability that motivated the original argument for putting assets on-chain.
Half the Gap Should Not Be Closed
Whether the composability gap is a defect depends entirely on the instrument. For a tokenized Treasury fund the restriction costs utility and protects nothing much; for a qualified-purchaser credit interest, unrestricted liquidation would defeat the exemption the fund relies on.
| Asset | Why it is restricted | Should the gap close? |
|---|---|---|
| Tokenized Treasury fund | Fund eligibility rules, often lighter | Yes — utility lost, little protected |
| Tokenized money market fund | Registration or exemption conditions | Largely — with an eligible liquidator set |
| Qualified-purchaser credit fund | Section 3(c)(7) exclusion depends on every holder qualifying | No — open liquidation breaks the exclusion |
| Reg D private placement interest | Accredited-investor conditions | No — only within the eligible set |
| Consumer credit receivables | Lending licences and consumer protection rules | No — issuer control is a legal requirement |
Three of the five rows say no, and they are not obstacles awaiting better engineering. An instrument whose exemption depends on every holder qualifying cannot have a liquidation path open to arbitrary bidders, because the liquidation would itself be the violation — the eligibility logic set out in distributed versus represented tokenized assets.
What Closing the Legitimate Half Requires
Enforcement that travels with the instrument. If eligibility is evaluated by the asset at the point of transfer, a protocol can hold and move it while the restriction still binds — including on a liquidation, provided the eligible liquidator set is defined in advance.
Required
- Eligibility checked at transfer, not at a platform edge
- A pre-defined set of eligible liquidators
- A price the protocol can rely on between NAV strikes
- A defined outcome when liquidation cannot complete
Insufficient
- Wrapping the asset to hide its restrictions
- Whitelisting only the protocol contract
- Assuming a liquidator will be found in time
- A daily NAV used as a live price
Ask before depositing
- Who can receive this on a liquidation?
- Has that path ever been executed?
- What price is used between NAV strikes?
- What happens if no eligible bidder appears?
The first item in the middle column deserves particular scepticism. Wrapping a restricted asset into an unrestricted token does not remove the restriction; it relocates the compliance failure to the wrapper and creates a claim whose enforceability depends on a structure the underlying issuer never approved. It looks like composability and is closer to an unhedged legal position — the rehypothecation dynamics covered in tokenized Treasuries as DeFi collateral.
How Blockmaze Makes Restricted Assets Usable
The requirement is that an asset keeps enforcing its own rules wherever it goes. That turns a restriction from a boundary a protocol must respect externally into a property the instrument carries, which is what allows collateral use without abandoning eligibility.
Eligibility Enforced at Transfer
The instrument evaluates every transfer against its own rules, so it can sit inside a protocol contract and still refuse an ineligible recipient — the property that makes controlled collateral use possible at all.
Liquidator Eligibility Declared
Which parties may receive an instrument on enforcement is recorded in advance, so a liquidation path exists and is testable before a default rather than discovered during one.
Encumbrance Recorded
Pledged and unencumbered states are properties of the position, so an asset cannot be posted as collateral in two places or counted as free while it is committed.
Valuation Source Recorded
The feed and timestamp behind a collateral value are retained, so a protocol relying on a daily NAV between strikes is doing so knowingly rather than treating a stale figure as live.
The second item is what converts a theoretical integration into a tested one. A protocol that can name every party eligible to receive collateral on a default has a liquidation mechanism; one that assumes an eligible bidder will materialise has an untested assumption sitting where its recovery process should be — the enforcement model described in smart contract compliance for RWAs on Layer-0.
Want Your Asset Usable as Collateral?
Blockmaze enforces eligibility at every transfer, declares who may receive an instrument on enforcement, records encumbrance, and retains the valuation source behind each collateral value.
Frequently Asked Questions
How much tokenized RWA value is actually used in DeFi?
Roughly $2.7 billion of about $27 billion in tokenized assets under management is actively deposited across DeFi lending markets — approximately 10%. Some measures put the figure lower still, at around $2.47 billion against nearly $30 billion tokenized. Either way the conclusion is the same: nine-tenths of tokenized value sits outside the composable financial system that tokenization was supposed to unlock.
Why can't permissioned tokens simply be deposited into DeFi protocols?
Because a lending protocol must be able to seize and sell collateral from a defaulting borrower, and a permissioned token restricts who may receive it. If a liquidator is not on the whitelist, the transfer reverts and the liquidation fails — leaving the protocol holding an asset it cannot sell against a debt it cannot recover. The conflict is structural: the transfer restriction that makes the token compliant is the same property that breaks the liquidation path.
How do the protocols that do accept RWA collateral solve this?
By becoming permissioned themselves at the point that matters. Aave Horizon, which had roughly $550 million in deposits by early 2026, operates as a dedicated market with participant controls rather than accepting restricted assets into the general pool. The pattern is consistent — the protocol accepts the asset's restrictions and maintains a set of eligible liquidators, rather than the asset shedding restrictions to fit an open protocol.
Which tokenized assets are actually being used as collateral?
Predominantly the most liquid and most standardised. BlackRock's BUIDL, at roughly $2.9 billion, is used across reserves, margin and lending, with institutional traders posting it as margin on Deribit. Janus Henderson's JAAA AAA CLO product and Apollo's ACRED credit fund also appear. Morpho on Avalanche reached around $875 million of RWA exposure. The concentration follows the same rule as everywhere else in this market: depth attracts depth.
Is the composability gap a problem or a feature?
Both, and which one depends on the asset. For a tokenized Treasury fund held by institutions that want to post it as margin, the gap is a genuine loss of utility with no offsetting benefit. For a restricted private credit interest that may only be held by qualified purchasers, the gap is the exemption working correctly — an asset that could be freely liquidated to an anonymous bidder would not remain compliant. Treating the whole gap as friction to remove misreads half of it.
What would close the gap without breaking compliance?
Enforcement that travels with the asset rather than living in a platform. If eligibility is evaluated at the point of transfer by the instrument itself, a protocol can hold and move the token while the restriction still binds — including on a liquidation, provided the liquidator set is defined in advance. That is the difference between an asset that cannot leave its platform and one that carries its rules with it, and it is the technical work that converts represented assets into usable collateral.
Related Articles
Is a Token You Cannot Move Really Tokenized?
The classification behind the gap — what can move versus what is recorded.
Why Do 56% of Tokenized Assets Never Move?
The same market measured by transfer activity rather than by DeFi deployment.
Tokenized Treasuries as DeFi Collateral and Rehypothecation Risk
What happens once tokenized collateral is reused across protocols.
Smart Contract Compliance for RWAs on Layer-0
Enforcing eligibility at the point of transfer rather than at a platform boundary.