[ARFC] Onboard HINC (Neuberger Securitize High Income Tokenized Fund) to Aave Horizon

[securitize]

[ARFC] Onboard HINC (Neuberger Securitize High Income Tokenized Fund) to Aave Horizon

Author: Securitize

Date: 2026-08-18

Target instance: Aave Horizon (Ethereum)

Asset: HINC (Neuberger Securitize High Income Tokenized Fund Ltd.)

---

Summary

We propose onboarding HINC, the tokenized share class of the Neuberger Securitize High Income Tokenized Fund Ltd., as supply-only collateral on Aave Horizon, with USDC, GHO and RLUSD borrowable against it.

HINC is a BVI professional fund advised by Securitize Capital LLC and sub-advised by Neuberger Berman Investment Advisers LLC (NBIA), an SEC-registered investment adviser and part of a group managing approximately $567bn in AUM as of 31 March 2026. The Fund invests the majority of its assets in high yield corporate bonds, with the balance in CLO tranches, bank loans and other high-yielding fixed income, plus a liquidity sleeve of cash equivalents and tokenized Treasury/MMF instruments.

Shares are issued as permissioned DSTokens (extended ERC-20 compatible tokens) on Ethereum under Securitize’s DS Protocol, with Securitize Transfer Agent, LLC (SEC-registered transfer agent) maintaining the Master Securityholder File. This is the same issuer stack already integrated with Aave Horizon for VBILL (VanEck Treasury Fund).

---

Motivation

Aave Horizon’s collateral set today is concentrated in tokenized Treasuries, government money market funds and investment-grade credit. That is the correct foundation, but it caps the economic depth of the venue: when collateral yields sit close to or below the stablecoin borrow rate, borrowing is a liquidity operation rather than a carry trade. Utilisation is therefore driven by episodic cash needs rather than by a persistent, rate-sensitive demand curve.

HINC changes that. As the first sub-investment-grade credit collateral on Horizon, it introduces an asset whose expected return sits materially above prevailing stablecoin borrow rates, which creates two distinct and durable demand sources:

1. Carry trade demand. Certain eligible Horizon users may independently elect to use HINC as collateral to borrow stablecoins and re-subscribe. Positive net carry over the borrow rate makes this a rate-responsive, repeatable strategy rather than a one-off draw. This is the demand profile that produces sustained borrowed TVL and reserve-factor revenue for the Aave DAO.

2. Balance-sheet financing. Credit funds, family offices and treasury desks holding HINC as a yield allocation gain a 24/7 liquidity line without triggering a redemption that would cost them the position and the carry.

Both of these are borrow-side demand, which is what Aave Horizon’s stablecoin suppliers are underwritten against. HINC aims to be additive to Aave Horizon’s collateral set rather than duplicative of existing collateral types.

Disclaimer: The use cases described are illustrative examples of potential protocol-level activity by independent eligible users and do not reflect the investment objectives, strategies, portfolio management activities, or recommendations of the Fund, Securitize Capital, the Transfer Agent, or the Sub-advisor. Any such activity is undertaken independently by users and is not managed, recommended, or facilitated by these parties.

On asset categorization

We want to address the categorization question directly, because it has been a blocker on prior RWA proposals. HINC is not a new asset category for Aave Horizon. It is a tokenized share of a professionally managed, NAV-priced fund similar to other tokenized funds like VBILL which is already onboarded. The delta is credit quality and volatility of the underlying portfolio, which is a risk-parameter question, not a category-approval question.

---

Asset Overview

Structure

Item	Detail
Fund	Neuberger Securitize High Income Tokenized Fund Ltd.
Domicile	British Virgin Islands business company; recognised as a professional fund under SIBA
Structure	Single-tier direct fund. No wrapper, no feeder, no master-feeder. Investors hold shares in the Fund itself
Investment Advisor	Securitize Capital LLC (Delaware; SEC-registered IA))
Sub-advisor	Neuberger Berman Investment Advisers LLC (SEC-registered IA, discretionary authority)
Transfer Agent	Securitize Transfer Agent, LLC (SEC-registered transfer agent)
Placement Agent	Securitize Markets, LLC (broker-dealer, FINRA/SIPC member)
Offering exemptions	Reg D 506(c) for U.S. persons; Reg S for non-U.S. persons. No exchange or ATS listing
Fund term commenced	August 18, 2026

Economic terms

Item	Detail
Management fee	0.50% p.a. of net assets, accrued daily, paid monthly in arrears
Total expense ratio	0.60% p.a. inclusive of management fee and operating expenses
Sub-advisory fee	23.75bps p.a., borne by the Investment Advisor, not the Fund
Initial share price	$1,000; subsequent issuance at NAV per share
Minimum subscription	$100,000 initial / $1,000 additional
Subscriptions	Fiat USD or USDC/USDG stablecoins, accepted each business day. Subscribing via USDC is atomic and instant via smart contracts
Distributions	None expected. Income is reinvested — HINC is a NAV-accruing, non-rebasing token
Lock-up	24 hours from subscription
Redemption	Requests any business day, 2:00pm ET cut-off. Fund targets T+1 for redemption payouts
Liquidity Pool Fee	An optional fee up to 2% of redemption value for instant redemptions. The liquidity pool is expected to be unavailable at launch (there are no fees for regular redemptions)
Liquidity sleeve	Cash, T-bills, commercial paper, repo, MMFs, tokenized Treasuries and fiat-pegged stablecoins

Tokenization

HINC tokens are digital representations of the underlying shares — not a separate security, and not tradeable independently of the shares. They are issued to six decimal places, held in self-custody in wallets approved through the Fund’s whitelisting procedures, and enforced at the token-contract level by DS Protocol:

• Allowlist enforcement in the transfer function. Only KYC/AML-cleared, subscribed wallets can hold or receive HINC. A non-whitelisted address cannot receive the token under any code path.

• UBO attribution. Every wallet balance maps to an identified beneficial owner in the Master Securityholder File.

• Transfer Agent Executive Actions. Freeze and seize capability, exercisable to comply with court orders, sanctions, or estate/lost-key events.

The practical consequence for Horizon’s stablecoin suppliers is worth stating plainly: because collateral cannot leave the allowlisted set, theft-style exfiltration of HINC collateral is not the operative risk. Illiquidity and NAV mark risk are. The remainder of this proposal is organized around those two.

---

Strategy and Risk Profile

The Sub-advisor targets attractive risk-adjusted returns through high yield bonds sourced in primary and secondary markets, with a CLO allocation expected to float between 0% and 30% over time. The Fund is USD-denominated and invests primarily in USD instruments.

Illustrative risk metrics

Below are derived from an illustrative index blend — 70% ICE BofA US HY Constrained / 30% J.P. Morgan CLOIE Post-BB, daily, constant-weight, 1 July 2016 to 15 July 2026. They do not reflect Fund-level fees and expenses, trading costs, defaults, realized losses, taxes or changes in market value. Actual portfolio composition and yield may differ materially. The Fund is newly launched or not yet operational, and the model portfolio does not represent actual holdings.

The 70/30 weighting is deliberately run at the maximum contemplated CLO allocation so the figures represent the strategy’s risk profile at fullest CLO utilisation, not the expected allocation on any given day. Risk providers should treat these as the conservative bound.

Metric	Illustrative Result
Annualised return since inception (2016–2026)	+7.21%
Worst month (March 2020)	−18.25%
Calendar year 2020	+7.75%
Calendar year 2022	−8.97%
Max drawdown within 2022	−13.20%
Effective / modified duration	2 – 3 years
Spread duration	3.5 – 4.5 years
Weighted-average life	5.5 – 7.5 years
Instantaneous NAV impact, +200bps spread widening	≈ −7% to −9%
Instantaneous NAV impact, +400bps spread widening	≈ −14% to −18%

Stress-period and volatility figures are hypothetical and based on the illustrative index blend, not actual Fund performance. Historical stress behavior may not predict how the Fund will perform in future market conditions.

An important distinction for parameterisation: the strategy is short interest-rate duration but carries meaningful spread duration. The CLO sleeve is floating-rate with near-zero rate duration. “Short duration” here should be read in the rate sense only — this is not a low-volatility asset, and the March 2020 figure should be treated as the governing stress case.

Concentration and leverage

• Corporate issuer concentration: generally contained within 5% per issuer.

• Sector concentration: generally within 20%.

• Fund-level leverage: none. The Fund does not borrow or employ leverage at the fund level.

• Embedded leverage: CLO debt tranches carry structural leverage of roughly 6–8x at the BB level. This is a driver of mark-to-market volatility. For context on realised impairment, cumulative BB CLO tranche defaults have historically run around 1.4%, an annualised rate near 0.04% — comparable to single-A rated corporates. Mark volatility and capital impairment are distinct here and should be parameterised separately: liquidation thresholds should be set against mark volatility; bad-debt expectations against impairment.

• Correlation note: HINC’s exposures overlap materially with Neuberger Berman’s other non-investment-grade credit strategies, which share the same team and research process. Allocators holding HINC alongside other NB non-IG products should treat those as correlated. There is little overlap with NB strategies outside non-IG credit.

---

Compliance Architecture

This section follows the RWA integration model established in the existing Securitize–Horizon integration.

Recordkeeping and UBO look-through

Tokens supplied to Horizon remain part of each investor’s legal holdings. Securitize Transfer Agent, LLCrecords them at UBO level in the MSF, flagged as encumbered collateral, and surfaces them in the investor portal as part of total securities held. The TA reconciles balances, encumbrances and lifecycle events through protocol APIs and subgraphs.

Horizon’s non-transferable aTokens satisfy the requirement that beneficial ownership remain tied to an allowlisted, KYC-verified wallet. Horizon does not need to register as an investor in HINC to operate the market — registration is only required for entities acting as liquidators.

Lifecycle and Executive Actions

Scenario	Protocol Requirement	Securitize Transfer Agent, LLC Action
Lost keys / death	Position migration: move full collateral and associated debt to a new wallet	TA updates MSF to new UBO wallet
Court order / sanctions	Freeze position; disable borrowing and liquidation for that account; notify liquidators the position cannot be unwound	TA freezes MSF entry, coordinates with regulators on resolution
Regulatory reserve freeze	Halt deposits/withdrawals on the reserve	TA documents and reports
Liquidation	Transfer the underlying HINC position to a pre-onboarded liquidator against stablecoin repayment	TA records liquidator as new holder of record
All of the above	Emit immutable events for MSF reconciliation	TA reconciles

Risk allocation on Executive Actions — an item requiring explicit governance acknowledgement

A Transfer Agent freeze or seize is a legal obligation, not a discretionary act, and it can leave a position temporarily un-liquidatable while its debt continues to accrue. That creates a defined path to bad debt that is not attributable to market risk, oracle failure, or protocol design.

We are proposing this be handled explicitly rather than left ambiguous:

1. Notice. Securitize Transfer Agent, LLC notifies Horizon operators and registered liquidators as promptly as legally permitted upon any Executive Action affecting a collateralised position.

2. Suspension of accrual for frozen positions, or an equivalent mechanism, so a borrower under a legal freeze does not accrue debt they are prohibited from servicing.

3. Explicit acknowledgement in the final AIP of how bad debt arising from an Executive Action is allocated between the Fund, the liquidator and the protocol.

We consider item 3 a prerequisite for go-live, in either direction. Silence on this point is the worst outcome for all parties.

---

Pricing and Oracle

Item	Proposal
Primary feed	Chainlink HINC NAV-link feed wrapped with LlamaGuard NAV dynamic bounds
NAV cadence	Daily, calculated each business day from the end-of-day net asset value
Feed type	NAV per share, USD
Growth-rate cap	CAPO-style adapter with a maximum annualised growth cap to bound feed manipulation and mis-publication. We propose 15% APR, providing headroom over the illustrative 7.21% while constraining the tail
Downside handling	No smoothing or floor. Negative NAV moves must pass through immediately — this is a credit asset and dampening the mark is precisely what would create bad debt
Staleness / heartbeat	Feed staleness threshold to be set by risk providers; borrowing pauses on stale feed
Emergency pause	Multisig authority to pause borrowing in the HINC market on feed anomaly

Liquidity and Liquidation Design

This is the section that matters most, and where NAV-priced credit differs fundamentally from crypto collateral.

Why the liquidation problem is different

For a crypto-native asset, liquidation is a race: price gaps intraday, and the backstop must be sized to absorb an instantaneous move against a market-clearing bid. For a NAV-priced credit fund, there is no intraday gap risk. NAV is struck once per business day. The mark moves in daily increments driven by spread and rate moves, not in seconds driven by liquidity cascades.

What the system requires is therefore not millisecond execution but reliable windowed liquidation — a guarantee that a position can be unwound within a bounded number of business days at or near the published mark. This materially changes required backstop capital. For a gradual-drawdown high yield fund, gross liquidation capacity should be sized as a percentage of borrowed TVL, not “sized to market.”

Sizing under the stress case: the illustrative worst month is −18.25%, which is roughly −0.9% per business day. Against a liquidation threshold set with meaningful headroom, and a 3-business-day liquidation window, the exposure to be covered is on the order of 3–5% of borrowed TVL — not a multiple of it. Ready liquidation capacity in that range, backed by committed liquidators, is the correct target.

Liquidator framework

Without pre-approved liquidators, this market cannot function — the underlying token only permits transfers between registered investors. Our commitments:

• No go-live before at least one liquidator is fully KYB’d, onboarded to HINC, and has stablecoin capacity committed. This is a a hard gate, not a launch-week task.

• Liquidators must be onboarded as holders of record of HINC by Securitize Transfer Agent, LLC in advance of any liquidation event, and must be able to both assume the asset and return stablecoins to the pool.

• The aim is for a minimum of two independent 3rd party liquidators before cap expansion, to avoid single-liquidator dependency.

• Liquidators receive Executive Action notifications from Securitize Transfer Agent, LLC and are contractually restricted from attempting to unwind frozen positions.

Disclaimer: Liquidators are independent third parties and are not affiliated with, controlled by, or acting on behalf of the Fund, Securitize Capital, the Transfer Agent, or the Sub-advisor. The availability, capacity, or participation of any liquidator may vary, and neither successful liquidation nor liquidity for HINC is guaranteed.

Redemption path and fee treatment

A liquidator’s ultimate exit is redemption at NAV through the Fund. Two frictions to state honestly:

1. Settlement timing. The Fund targets T+1 for redemption payouts. While the corporate bonds trade publicly in a highly liquid market; bid-ask can widen materially in stress. Liquidators must be capitalised to carry the position across the settlement window to avoid undertaking pro-longed holding risk..

2. Liquidity Pool Fee. Optional liquidity pool instant redemptions are offered up to 2% of redemption value. This would otherwise be a direct drag on liquidation economics. This optional liquidity pool is expected to be unavailable at launch; regular redemptions do not incur any fees.

---

Technical Asset Listing Framework Alignment

Framework Area	HINC Status
ERC-20 compatibility	Extended ERC-20 with allowlist-enforced transfers under DS Protocol. Six-decimal fractional issuance. No rebasing, no fee-on-transfer
Oracle design	Chainlink NAV feed for HINC, published under the TSSO standard
Access control / privileged roles	Transfer Agent holds allowlist, freeze and seize powers. Investment Advisor may cancel and reissue tokens. Fully disclosed; these are regulatory requirements, not discretionary admin keys. Role holders and multisig topology to be published in the technical assessment
Mint / burn logic	Minting on accepted subscription; burning on redemption. Both gated by the Transfer Agent
Pause / blacklist	Present by design and regulatorily mandated. Governance should treat these as a documented feature, not a hidden risk
Upgradeability	To be documented in the technical assessment package
Exchange rate / yield mechanism	NAV-accruing, non-rebasing. NAV struck each business day at end-of-day net assets
Token architecture	Native issuance using DSToken architecture. No bridge exposure, therefore no bridge risk and no cross-chain supply reconciliation risk.
Audit history	Audit reports for the HINC token contracts and DS Protocol to be provided
External dependencies	Redstone (NAV feed); Fund Administrator (NAV calculation); Sub-advisor (portfolio valuation inputs); Custodian

Securitize Commitments

1. Securitize Transfer Agent, LLC: Full technical assessment package: contract addresses, audit reports, privileged-role inventory

2. Securitize Transfer Agent, LLC: Liquidator onboarding executed as a pre-launch gate, not a post-launch task.

3. Securitize Transfer Agent, LLC: Executive Action notification protocol to Horizon operators and registered liquidators.

4. Securitize Capital LLC: Ongoing publication of AUM, portfolio composition, and NAV history to support risk-provider monitoring.

5. Securitize I, Inc: Affiliated parent company of Securitize Capital, LLC will supply $1M in seed collateral at launch to initialize and establish the market.

   ---

Risks and Known Open Items

We would rather surface these than have risk providers find them.

• No operating history. The Fund’s term commenced 20 July 2026. All risk figures presented are index-derived illustrations.

• Liquidity mismatch. Investors face a 24-hour lock-up and daily redemption requests against a portfolio of HY bonds and CLOs that may require several days to liquidate. The 24-hour lock-up does not reflect practical liquidity. In stress, redemptions may be delayed, suspended, or paid below the NAV at request date.

• Sub-investment-grade credit risk. Unsecured and subordinated HY exposure with historically low recovery in default. Secondary market liquidity in HY deteriorates materially in dislocation.

• CLO structural risk. Overcollateralisation test breaches divert cash flow from junior to senior tranches; the Fund has no standing in portfolio-company insolvency proceedings.

• No replication of the Sub-advisor’s investment process by the Investment Advisor. The Investment Advisor does not independently replicate or duplicate the Sub-advisor’s investment research, security selection, or portfolio management processes.

• Token cancellation and reissuance authority. The Investment Advisor may cancel and replace tokens without notice, including for technical upgrades or theft response. Operationally disruptive if exercised while positions are collateralised; we will propose a coordination protocol.

• Suspension powers. Directors may suspend NAV determination, subscription or redemption.

• Oracle attestation chain not yet fully documented (see Pricing section).

• Chainlink feed requirement unresolved (see Pricing section).

---

Disclosure

Securitize is the tokenization platform, transfer agent and investment advisor for HINC and is the author of this proposal. We have a direct commercial interest in this listing. We have attempted to present risk data at its conservative bound and to disclose open items rather than defer them; risk providers should nonetheless treat this as an issuer-authored proposal and diligence accordingly.

Securitize Markets, LLC, member FINRA/SIPC, acts as placement agent for the Fund’s private placement. Nothing in this proposal is an offer to sell or a solicitation to buy any security. HINC is offered only to professional investors and qualified purchasers under Reg D 506(c) and Reg S.

Securitize Capital LLC is an SEC registered investment adviser. SEC registration does not constitute and endorsement from the SEC, nor does it imply a certain level of skill or training.

For avoidance of doubt, neither the Fund, Securitize Capital LLC, the Sub-advisor, nor the Transfer Agent operates, controls, manages, administers, or supervises the Aave protocol, Horizon markets, user borrowing activity, liquidation decisions, leverage strategies, or other DeFi activities undertaken by third-party users. Any borrowing, lending, collateralization, liquidation, looping, or similar activities are conducted solely by protocol participants pursuant to the rules of the applicable protocol.

---

Next Steps

1. Publication of this ARFC for community and service-provider feedback.

2. Technical assessment under the Technical Asset Listing Framework; resolution of the Chainlink feed and oracle attestation open items.

3. Risk assessment by LlamaRisk.

4. Liquidator onboarding completed and evidenced.

5. If feedback is constructive, escalation to ARFC Snapshot.

6. If Snapshot is YAE, publication of an AIP for final confirmation and enforcement — including explicit treatment of Executive Action bad-debt allocation.

---

Copyright

Copyright and related rights waived via CC0

[aksenovd10]

Same disclosure as on the mWIN Horizon thread: no commercial relationship with Securitize, Neuberger Berman, Aave Labs or LlamaRisk. I mark illiquid private positions for a living and I read these as a stablecoin supplier would.

Worth saying first that this proposal answers most of what I asked over there before anyone asked it. The liquidator gate is pre-launch rather than post-launch, the mismatch between a 24-hour lock-up and a multi-day high-yield unwind is disclosed rather than buried, and the Executive Action problem is raised by the issuer rather than by a commenter. That last one is the most important section in the post and I have not seen it picked up.

The backstop number is the one I would push on. The proposal sizes liquidator backstop capital at 3-5% of borrowed TVL against a three-business-day liquidation window. That is a clear, falsifiable commitment, which is more than most proposals offer, so it deserves to be tested rather than accepted.

The risk that backstop absorbs is not a fraction of borrowed TVL – it is the NAV move across the liquidation window plus the T+1 settlement leg, against whatever headroom the LLTV leaves. The disclosed risk figures are monthly: worst month -18.25%, 2022 maximum drawdown -13.20%, and -14% to -18% for a 400bps spread widening. A liquidation window is four business days, not a month, so none of those speaks to the exposure actually being covered.

The figure that speaks to it is a four-day one, and there are two of them. Over the same July 2016 to July 2026 blend: what is the worst four-business-day fall? And what is the worst four-business-day fall that begins only after the blend has already dropped far enough to put a position into liquidation? The second is the one that sizes the backstop, because liquidations do not begin on quiet days. They begin part-way into a bad run, and bad days cluster. Sizing a backstop at 3-5% of borrowed TVL is defensible if those numbers are small. If the worst four-day move inside March 2020 was materially larger than five per cent, the backstop is sized below the event it exists for. The blend is already built, so both figures should be cheap to produce.

There is also an open thread on precisely this question. GBQuant’s independent liquidation-capacity work, posted a day after this proposal, models liquidator warehousing economics explicitly – funding cost, capital hurdle, hedge cost, execution loss and redemption throughput – and arrives at a required bonus rather than assuming one. It does not cover RWA collateral or permissioned liquidators, and it asks whether the risk framework should supplement instant clearance with a specified exit horizon and stressed primary-redemption throughput. This proposal is a direct instance of that question: permissioned liquidators, no secondary market, exit only by NAV redemption through the fund. The two threads belong together, and a 3-5% figure derived on that basis would be far more convincing than one asserted.

The two live RWA onboardings have oracles asymmetric in opposite directions. Here, a CAPO-style adapter caps NAV growth at 15% APR while downside passes through immediately with no smoothing or floor. On the mWIN thread, a static band sits 0.8% below the issue value and 15.4% above. One design bounds the direction that cannot hurt a lender and lets the other run; the other bounds the direction that can hurt a lender very tightly and lets the upside run.

Both are defensible in isolation. I do not think both can be the house view. For an asset whose own disclosure puts a -18% month inside the sample, the absence of any downside floor is a deliberate choice with a real consequence: a single inaccurate NAV print marks the entire market down at once, and every position that breaches its threshold enters the same three-day liquidation window at the same moment. Is the intended rule that RWA feeds bound manipulation upward only, and that downside protection lives entirely in the LLTV and the liquidation bonus? If so, stating it once in the framework rather than per asset would settle a whole class of repeatable questions.

On Executive Actions, the open side of the ledger. The proposal asks for AIP-level acknowledgement of how bad debt from an Executive Action is allocated between the Fund, the liquidator and the protocol, and proposes suspending accrual for frozen positions so a borrower under a legal freeze does not accrue debt they are prohibited from servicing. Both are right, and both sit on the borrower’s side of the ledger. The stablecoin supplier is on the other: funding capital that is illiquid, un-liquidatable and no longer earning, for a duration set by a court rather than a market. Does the proposed allocation contemplate compensating that supplier, and out of what?

Three smaller things, none of them risk questions. The oracle provider is named twice and differently: the primary feed is described as a Chainlink HINC NAV feed wrapped with LlamaGuard dynamic bounds, while the external dependency list names Redstone as the NAV feed, and the Chainlink feed requirement is separately noted as unresolved. Which is it, and does the answer change the attestation chain?

The bounds here are described as dynamic, where the mWIN feed’s are static values set at deployment. What makes them dynamic, what is the input, and who holds the authority to move them?

And the fund’s term is given as commencing 18 August 2026 in one place and 20 July 2026 in another – a month either way matters more than usual here, since operating history is the one thing this collateral has none of.

None of this is opposition to onboarding, and the disclosure standard here is above the norm. The backstop figure is the single item I would want derived properly rather than stated before this reaches a vote.

[GBQuant]

Thanks for connecting the two threads. I ran the four-session statistics you asked for on a public proxy, and committed the code, the data, and a manifest so anyone can re-run them.

Two things came out of it. First, “3-5% of borrowed TVL” answers a different question depending on what it means. Second, on the proposal’s literal gross-capacity reading, the missing test is simultaneous repayment rather than the NAV move in isolation.

The phrase covers three different quantities

1. gross stablecoin financing available to repay debt and receive HINC;
2. loss-absorbing capital for the NAV move between liquidation and redemption;
3. the liquidation bonus that compensates funding, settlement delay, recovery
   loss, and the liquidator’s cost of capital.

The proposal describes 3-5% as gross liquidation capacity relative to borrowed TVL. That is the first one.

On that definition, the constraint is concurrency

Gross capacity of 3-5% is sufficient only if no more than 3-5% of borrowed TVL needs repaying at the same time. Two features of the proposal make that the hard part.

The NAV feed is a common trigger. One daily mark moves positions through their thresholds together, not one at a time: a 5% print catches every position with less than a 5% health buffer. How many that is depends on the distribution of LTV and health factors across the book, which is not published.

The same capital also cannot be assumed to recycle immediately. Unless earlier redemptions settle or separate credit is available, stablecoins committed to a liquidation remain tied up until settlement. Treating 3-5% as a one-shot ceiling rather than daily throughput is therefore the conservative interpretation; the exact recycling schedule depends on redemption timing and cut-offs.

So: if the capacity is gross, cannot be assumed to recycle before settlement, and the trigger is common to the whole book, what keeps less than 5% of the book from crossing the threshold on the same day? That question needs no NAV series to ask, only the position-level distribution to answer.

On the economic reading, the threshold inverts

If instead 3-5% means compensation per dollar of debt repaid, the warehouse equation can be inverted without any HINC or proxy data. With 10% annual funding, a separate 10% capital hurdle, and six calendar days from liquidation to cash recovery:

economic compensation   maximum supported NAV loss
3%                                  2.59%
5%                                  4.45%

Read this way, 3-5% works if and only if the recovery loss stays below 2.59-4.45%. That ceiling follows from the cost assumptions alone, and is a different quantity from the stablecoin commitment unless the same percentage is meant to do both jobs.

What a public benchmark says about that ceiling

I added a reproducible RWA drawdown analysis, with a pinned manifest and the committed series: 2,522 Yahoo Finance HYG adjusted closes from July 1, 2016 to July 15, 2026,
retrieved August 31, 2026, with four-session returns taken close-to-close across five observations.

This is not a HINC estimate. The exact daily 70/30 blend is not attached to the proposal, and HYG is a public market benchmark rather than HINC NAV or the J.P. Morgan CLOIE Post-BB component. It is useful only for judging whether a 4.45% ceiling is a large number or a small one for stressed high-yield credit.

condition                 start         end      prior drawdown   forward return
unconditional             2020-03-13  2020-03-19       n/a            -10.87%
after 5% drawdown         2020-03-13  2020-03-19      -9.05%          -10.87%
after 10% drawdown        2020-03-17  2020-03-23     -13.25%          -10.12%

The benchmark, which does not reproduce the proposed 30% BB CLO sleeve, lost 10.87% in its worst four sessions: roughly two and a half times the 4.45% ceiling.

You were right that liquidations start part-way into a bad run, since the worst window already begins 9.05% below its recent high. But conditioning does not produce a worse tail here. At the 10% condition the worst forward window is milder, 10.12% rather than 10.87%. That result holds for every rolling-maximum lookback from 20 through 250 sessions.

A transparent bracket for the blend

HYG lost 10.87% over those four sessions against 10.03% over the whole of March 2020, which is also its worst month in the sample. Its short-window loss was therefore 1.08 times its worst-month loss: the drawdown happened in days and then partly retraced.

Applying that concentration to the 18.25% worst month the proposal discloses gives a 19.78% four-session bracket. It assumes that HINC stress concentrates in time the way HYG stress did, which still needs testing against the exact blend.

What those losses cost a liquidator

For a normalized $100 million debt repayment, lump redemption at the end of the window, 10% annual funding, a separate 10% annual capital hurdle, no hedge, and no redemption fee:

scenario                         NAV loss   calendar days   minimum bonus
3% loss sensitivity                3.00%          4             3.32%
5% loss sensitivity                5.00%          4             5.49%
HYG worst four-session window     10.87%          6            12.56%
stress-shape bracket              19.78%          6            25.07%

The March 13 to March 19 window holds four trading sessions but six calendar days, because it crosses a weekend. Charging four days gives 12.44% instead of 12.56%, a 12 basis-point difference. And none of these percentages is repayment capacity: the liquidator still needs about $100 million at time zero to repay $100 million of debt, whether the recovery loss is 3% or 20%.

Permissioning makes both tests harder to pass

Aave fixes the bonus ex ante in either design, but an open market at least lets another profitable liquidator enter. A whitelist removes that fallback, so the economics and committed financing both have to be adequate for the approved firms before stress begins.

What would make this an actual HINC result

• the dated daily level or return series for the exact 70/30 blend;
• proposed LTV, liquidation threshold, liquidation bonus, and close-factor rule;
• the precise definition and dollar amount of the 3-5% commitment;
• the position-level debt and health-factor distribution, or a conservative simultaneous-trigger assumption;
• confirmation that three business days plus T+1 runs from debt repayment to cash in hand;
• redemption cut-off, throughput, fee, gating, and suspension assumptions.

With those, the same runner replaces the benchmark with the unconditional and trigger-conditioned statistics for the real blend, the simultaneous financing requirement, and the minimum bonus under normal and stressed settlement. Until then, 3-5% is a reasonable hypothesis, but it has not been tested against either of the two things it might mean.

---

Independent research, unaffiliated with Aave Labs, the DAO, or any Aave service provider. This is not a parameter recommendation and does not substitute for service-provider analysis. Everything above is reproducible from the linked manifest, and I would rather have the assumptions challenged than accepted.

[aksenovd10]

That is a better question than mine, and one of my two premises did not survive it.

You confirmed the first half: the worst window does begin part-way into a bad run, 9.05% below its recent high. The second half, that conditioning therefore produces a worse tail, is not supported by your data, and it fails across every lookback you tested. I asked for the conditional figure because I assumed it would be worse. It is not, and the unconditional four-session number is the one that matters and makes sense then.

The reframing from magnitude to concurrency is useful. What keeps less than 5% of the book from crossing the threshold on the same day needs no NAV series to ask, and a single daily mark is exactly the mechanism that makes simultaneity the default rather than the tail.

One challenge back, since you asked for them. Your 19.78% bracket applies HYG’s concentration ratio of 1.08 to the disclosed worst month, and you flag that HYG carries none of the 30% BB CLO sleeve. I would expect that sleeve to push the ratio up rather than down – mezzanine CLO in March 2020 moved later and further than cash high yield – so 1.08 reads to me as a floor for the blend rather than a point estimate. Worth testing if the CLOIE Post-BB series is obtainable.

One addition to your list of what would make this an actual HINC result, from the other proposal in this pair.

These two fail in opposite places. Here there is no downside floor, so a move on the scale of your bracket prints in full and the whole constraint sits in the liquidation machinery you have just modelled. On mWIN it is the reverse: minExpectedAnswer sits 0.8% below the $130,000 Authorised Denomination, so a move of that size cannot be reported at all. That band can only express its own worst disclosed drawdown once NAV has accrued roughly 7.6% above issue value, and only until it reaches the 150,000 ceiling.

That bears directly on your concurrency test. On mWIN the common trigger cannot fire, because the print that would move the book through its thresholds is outside the band before it gets there. Depending on which layer catches it – Horizon’s documentation describes DON-level rejection with the last valid price retained, while the Midas data feed contract reverts – the book either runs against a stale mark or the market stops. Neither state liquidates. So a seventh item, at least for the mWIN side: which layer governs a bounds breach, because it decides whether the liquidation economics are ever reached.

Thanks for committing the manifest. Appreciate the ongoing discussion here.

[GBQuant]

Thanks, this sharpens both points.

On 1.08, I agree that it should not be treated as a point estimate, but I would
not yet call it a floor. A purely multiplicative increase in severity affects
both the four-session and monthly losses; what changes their ratio is the shape
and timing of the path. A drawdown crossing a month boundary can raise the
ratio, while a late selloff captured by the month-end close can lower it. HYG’s
1.08 is consistent with the partial recovery before month-end.

At the blend level, asynchronous HY and CLO drawdowns can also reduce the ratio
because no single four-session window necessarily contains both legs’ worst
moves, while the calendar month can contain both. For illustration, a 70/30
blend in which the first leg loses 10% in sessions one to four and the second
loses 20% in sessions fifteen to eighteen has a 13% monthly loss but a 7% worst
four-session loss, giving a ratio of 0.54. Whether this dilution or the
month-boundary effect dominates is exactly what the daily blend would settle.

J.P. Morgan calculates CLOIE on each US business day, but the historical series
appears to be distributed through DataQuery, Bloomberg, or subscription rather
than as a public dataset. Since the proposal has already constructed the daily
70/30 blend, publishing the worst four-session result and its five dated levels
would close the test without requiring release of the full series.

I also agree that the mWIN observation identifies a prerequisite to the
liquidation analysis: oracle reachability. The Midas feed setter rejects an
out-of-bounds update, so no new round is stored and the latest valid round
remains readable. Both the DON path and the contract path therefore converge on
the same observable state, a readable stale round, which moves your seventh
item downstream rather than settling it at the feed layer. What happens next
depends on the heartbeat, staleness, adapter, and reserve controls. The
operational questions are: after a rejected update, what value and timestamp
does Horizon read, when is that value rejected as stale, and what automatic
pause or freeze follows?

That leaves four tests in sequence: can the oracle represent the stress, how
much debt triggers simultaneously, can approved liquidators finance that
repayment, and does the bonus compensate them for settlement and recovery risk.
My analysis has addressed the last test and framed the inputs required for the
middle two; oracle reachability remains a separate prerequisite.

[aksenovd10]

Thanks, conceded on 1.08 – the path argument is right, and the asynchronous-legs case shows the ratio can fall below one for a blend, so “floor” was wrong. The daily 70/30 blend settles it, and I would second the ask: the worst four-session result and its five dated levels would close the test without releasing the full series.

Agreed that the seventh item moves downstream. Both paths leave a readable stale round; what matters is what Horizon reads after a rejected update and when it freezes. One cross-reference: on the mWIN thread, Midas confirmed the Aave-side feed and its staleness parameters are LlamaRisk-defined, separate from the issuer’s own bounds. So your three operational questions are LlamaRisk’s and Horizon’s to answer, on both assets.

Your four tests in sequence read as the proper checklist for this pair then.