Tokenized real-world assets rarely finance at 100 cents on the dollar. They borrow below face value because the structures that make them safe for lenders also subordinate a portion of value and demand liquidity premia. The haircut is not a crypto quirk. It is securitization mechanics, risk policy, and market depth showing up on-chain.
Verified: Centrifuge’s Tinlake pools split collateral into a senior DROP token and a junior TIN token with an explicit first-loss buffer. The New Silver 2 (NS2) term sheet lists a minimum 20% junior risk buffer and targets a 7% DROP yield, so only about 80% of pool value is senior-backed at any moment, by design (NS2 executive summary). MakerDAO codifies this conservatism at the protocol level: its July 24, 2023 governance poll for New Silver shows a “Minimum Structure Subordination” of 20% and a 100% haircut on defaulted pledged assets (Maker poll).
Verified: Underwriting further restricts lendable value before tokenization. REIF1 caps first-position loans at no more than 70% of third-party appraised value, with seconds up to 80% (REIF1 executive summary). Then tranching subordinates additional value to TIN. The result is materially less than 100% collateral value to borrow against, even though the underlying loans have higher face value.
Timeliness: institutions are now plumbing tokenized RWAs into collateral workflows. BlackRock’s BUIDL, a tokenized short-term Treasury fund with roughly $2–2.6 billion AUM in 2026, gained a framework with OKX and Standard Chartered on April 28, 2026 to use BUIDL as yield-bearing collateral with bank custody (Standard Chartered press release). Meanwhile, tokenized Treasuries reached roughly $11–13 billion by March–May 2026, though the entire RWA segment remains small next to traditional markets (CoinGecko summary citing RWA.xyz). Yet thin on-chain liquidity persists: around 56% of tokenized RWA value showed no weekly on-chain transfer activity as of May 2026 (FinanceFeeds).
What materially changed to make haircuts unavoidable
Verified: RWAs are no longer purely experimental. Products like BUIDL are integrated as collateral in permissioned arrangements with exchanges and custodians (Standard Chartered). On the decentralized side, Maker’s vaults already accept senior DROP tokens, with protocol-level limits and haircuts enforced by governance (Maker poll) and by how Tinlake mints DROP and TIN (Tinlake docs).
Inference: As RWAs scale and move into collateral roles across venues, lenders apply the same tools they use in traditional markets to protect against losses and illiquidity. Subordination, conservative LTVs, and punitive treatment of defaulted assets are standard. Tokenization does not erase those constraints. It exposes them on-chain.
Tranches, LTV caps, and liquidity gaps: the strongest evidence
Verified: The structures and parameters that determine how much value lenders can safely advance are explicit in public documentation. The following datapoints come directly from pool term sheets and governance records, plus market trackers:
Driver
Example metric
Source
First-loss subordination
Minimum 20% TIN buffer (NS2)
NS2 executive summary
Underlying LTV constraint
First-position loans ≤ 70% of appraised value (REIF1)
REIF1 executive summary
Protocol risk policy
Defaulted pledged assets: 100% haircut (Maker RWA002-A)
Maker poll
Market liquidity
~56% of tokenized RWA value had no weekly on-chain activity
FinanceFeeds
Institutional adoption
BUIDL integrated as collateral framework with OKX and SC
Standard Chartered
Inference: Combine a 20% junior buffer with underlying loans capped at, say, 70% LTV and the immediately financeable senior layer is well below the face value of the collateral pool. Add thin secondary markets, and prudent lenders demand even more cushion or tighter advance rates.
Verified: Tinlake’s design explicitly mints the junior and senior claims, and Maker only accepts the senior DROP as collateral with a debt ceiling and conservative parameters (Tinlake docs), (Maker poll). That means protocols are taking a fraction of pool value by policy, not because the tokens malfunction.
Implications for DeFi lenders and borrowers
Verified: Maker’s RWA vaults reflect a cautious stance with “Minimum Structure Subordination” at 20% and a 100% haircut on defaulted pledged assets in the New Silver context (Maker poll). Tinlake’s NS2 sets the junior buffer at a minimum 20% and targets a 7% DROP yield (NS2 executive summary).
Inference: For stablecoin issuers and money markets, these settings cap leverage and underpin yields. Senior lenders earn returns commensurate with the subordination and liquidity risk. Borrowers funding against RWAs on-chain should expect lower advance rates than the headline value of their assets and tighter covenants as pools scale or performance varies.
Opinion: This is healthy discipline. By enshrining first-loss capital and punitive default haircuts, DeFi avoids importing the worst habits of pre-2008 securitization. The trade-off is lower capital efficiency for RWA-backed borrowing relative to highly liquid, rehypothecable crypto collateral. Over time, competitive pressure and better data could compress spreads, but the core haircut logic will remain as long as pools rely on junior capital and off-chain recoveries.
Institutional and market-structure effects
Verified: Tokenized Treasuries scaled to roughly $11–13 billion by March–May 2026, with the broader RWA market in the tens of billions, according to industry aggregation summarized by CoinGecko (CoinGecko). BlackRock’s BUIDL reached roughly $2–2.6 billion AUM and is now eligible as yield-bearing collateral at OKX within a framework involving Standard Chartered custody (Standard Chartered).
Inference: In permissioned venues where settlement, custody, and legal enforceability are bank-grade, some haircuts may narrow, particularly for short-duration, government-backed exposures. Operational and counterparty risks fall when a global custodian sits between token holders and the issuer. That does not eliminate duration, liquidity, or market risk, but it can reduce the additional discount applied purely for on-chain frictions.
Verified: Despite growth, liquidity remains patchy. Around 56% of tokenized RWA value had no weekly on-chain activity as of May 2026 (FinanceFeeds).
Opinion: The split market will persist for a while. Blue-chip tokenized Treasuries in bank-custodied frameworks are likely to command the tightest haircuts. Long-tail private credit and real-estate pools will keep wider discounts until they demonstrate steady performance, auditable cash flows, and reliable secondary liquidity.
The strongest counterargument: haircuts can compress fast
Counterpoint: If tokenized Treasuries keep scaling and trade with visible depth, lenders may treat them like traditional repo collateral with modest haircuts. The BUIDL framework with OKX and Standard Chartered is a concrete step toward treating tokenized fund shares as operationally robust collateral (Standard Chartered), and the broader market already sits in the low tens of billions (CoinGecko).
Rebuttal: Even if operational risk shrinks, structural subordination remains. Tinlake pools still allocate a first-loss TIN tranche and impose LTV caps for the underlying loans (NS2), (REIF1). Maker’s parameters explicitly recognize that defaulted pledged assets get a 100% haircut (Maker poll). And the liquidity gap is real: more than half of tokenized RWA value saw no weekly on-chain activity (FinanceFeeds). In other words, haircuts can compress at the margin but cannot vanish where junior capital and illiquidity must absorb losses.
What would confirm or weaken this thesis
- Verified disclosure cadence: New Tinlake pool summaries showing lower minimum TIN ratios or higher senior advance rates would signal haircut compression; the reverse would confirm caution. Watch future executive summaries and term sheets (Tinlake docs).
- Governance parameters: MakerDAO votes adjusting “Minimum Structure Subordination,” default haircuts, or vault debt ceilings for RWA collateral will reveal appetite for risk or conservatism (Maker governance).
- Liquidity metrics: A declining share of tokenized RWAs with no weekly on-chain activity would weaken the case for large liquidity premia; persistence of the ~56% figure would support it (FinanceFeeds).
- Institutional frameworks: More arrangements like the OKX–BlackRock–Standard Chartered model, including published margin schedules and eligible collateral lists, would test whether permissioned settings materially narrow haircuts (Standard Chartered).
- Market scale and breadth: Continued growth of tokenized Treasuries toward the high tens of billions, plus diversification beyond government paper, would indicate maturing depth. Stagnation would keep haircuts elevated (CoinGecko).
- Recovery outcomes: Documented recoveries in default scenarios within RWA pools, and how they flow through TIN before DROP, would validate the protective function of subordination and inform future advance rates (NS2), (REIF1).
Bottom line: tokenized collateral borrows below par because securitization math, credit policy, and liquidity realities demand it. As institutional rails harden and secondary markets deepen, some discounts may narrow. The structure that protects lenders will still take the first bite out of face value.
Disclaimer: This article is provided for informational purposes only. It is not offered or intended to be used as legal, tax, investment, financial, or other advice.
