Driving Morpho's IRM to a Locked 8% APY

The situation, stated plainly

Right now the UnblockEquity genesis vault on Base has roughly $16K of TVL and is displaying a combined APY of about 65% APR. A reasonable response from a DeFi-native reader is: that's either a scam, a fire sale, or someone is about to learn an expensive lesson about mercenary capital.

It's none of those. It's a deliberate, time-bounded reward bridge designed to attract the first wave of external liquidity providers while we drive Morpho's AdaptiveCurveIRM toward a specific destination: a permanently locked 8% net depositor APY at exactly 90% utilization. The math says it takes about 8 months at sustained ~99% util. This post walks through that math engineer-to-engineer.

Where we started — a genesis loan, not a yield farm

UnblockEquity is a tokenized junior-lien home-equity protocol. The first borrower is the founder — me. The collateral asset is an ERC-20 called eqBLOCK, minted 1:1 against tokenized home equity at a 62.5% LLTV against an oracle-fed AVM. The borrower takes USDC out of the Morpho market against that collateral. The depositor side is a Morpho vault that allocates to that market.

The reason this protocol exists at all: HELOC-denied homeowners cannot access their own equity through any traditional channel. Banks won't lend if you've missed payments in the last 12 months — the exact moment you need equity access most. The protocol replaces the bank's denial with on-chain collateralization. The first deal is mine because I'm in pre-foreclosure on my own house and I built the product I needed. That's not pathos; it's skin in the game.

So when we say “we will execute a matching borrow against eqBLOCK collateral within 24 hours of every external LP deposit,” that's not an aspirational roadmap. It's the founder paying his own mortgage with each new dollar of vault liquidity. The incentives are aligned because there is no escape valve.

The IRM problem (and the opportunity inside it)

Morpho Blue markets use the AdaptiveCurveIRM. The relevant mechanic, simplified to one sentence: at 90% utilization the rate curve sits exactly at rateAtTarget, and rateAtTarget itself drifts up or down based on how far the current util is from 90%.

Crucially, the drift is multiplicative and exponential in time, not additive. Sustained util above 90% multiplies rateAtTarget upward at a rate proportional to the distance from target. Sustained util below 90% multiplies it downward.

Our market spent weeks at ~86% utilization while we bootstrapped the contracts, fixed indexing issues, and waited for Morpho's API to whitelist the V2 factory. During that period rateAtTarget bled from an initial ~4% APR down to about 0.35% APR. That's why the displayed base APY hit 0.31% the day before this post. The IRM was doing exactly what it was designed to do: telling the market that nobody seemed to want our credit.

  rateAtTarget over the last 60 days (illustrative)

  4.0% APR ┤●
           │ ●●●
           │    ●●●
  2.0% APR ┤       ●●●●
           │           ●●●●●
           │                ●●●●●●●
  1.0% APR ┤                       ●●●●●●●
           │                              ●●●●●●●●●
  0.35%    ┤                                       ●●●●●●● ← today
           └──────────────────────────────────────────────
            Genesis            ~86% util sustained        Now

The mechanism we found — drift is bidirectional

The same math that punished us at 86% util can reward us at 99% util. The AdaptiveCurveIRM doesn't care which direction it's pushing — it cares about the distance from 90% and the time spent there.

Below is the IRM's drift speed as a function of cap utilization, and the time required to grow rateAtTarget from today's 0.35% APR to the 9.91% APR we're targeting.

99% is the sweet spot. Going higher buys diminishing returns and costs LP exit buffer. So 99% is where the webapp cap is configured today.

The math of lock-in

At exactly 90% utilization, the IRM's drift term is zero. rateAtTarget stops moving in either direction. This is the load-bearing fact in the entire strategy.

So if we can drive rateAtTarget up to ~9.91% APR during a high-util bootstrap phase, then ratchet utilization down to exactly 90%, the IRM holds the rate in place. From there the depositor APY math is:

APY_depositor = rateAtTarget × curve_mult × util × (1 − perf_fee)

At util = 0.90:
  curve_mult = 1.00          (exactly at target)
  perf_fee   = 0.10          (10% to protocol)
  APY = rateAtTarget × 0.90 × 0.90
      = rateAtTarget × 0.81

For APY_target = 8.03%:
  rateAtTarget = 8.03 / 0.81
               = 9.91% APR

8.03% is not an arbitrary number. It's the risk-priced target we modeled for the BR12 (12-month escrow) product class — in line with private-credit and bridge-loan funds backed by junior real-estate liens. It's the number a curator would accept as “reasonable yield for the stated risk.”

The match-borrow commitment

Driving util to 99% is easy when you're the only depositor and the only borrower. The hard part is keeping util at 99% as external LPs arrive. The mechanism: every external LP deposit gets matched within 24 hours by a corresponding borrow against eqBLOCK collateral from the founder wallet.

This is auditable on-chain. Every Deposit event on the vault should be followed by a Borrow event from the founder wallet within 24 hours. We publish a monthly compliance report. To be precise: this is a behavioral commitment, not a smart-contract-enforced rule. The reason it's credible is that the borrowed USDC funds my actual mortgage arrears, so my incentive is perfectly aligned with the strategy.

  External LP deposits X USDC
        ↓
  Founder borrows X USDC (within 24h) against eqBLOCK
        ↓
  Vault utilization maintained at ~99%
        ↓
  IRM drift = +5/year toward rateAtTarget = 9.91% APR
        ↓
  Borrowed USDC → FundDistributor → arrears bucket
        ↓
  Wire to mortgage servicer (SPS) → loan rehabilitation
        ↓
  Better story, more curator interest, more LP deposits
        ↓
  (repeat for 8 months)

The reward bridge

8 months is a long time to wait for base APY to grow. The bridge is a Merkl campaign — a 31-day USDC reward distribution to vault depositors. The first tranche is $500 over 30 days. At today's $16K TVL that prints a headline reward APR of roughly 65%. It is meant to be eye-catching. It is also meant to dilute as TVL grows.

Reward subsidy shrinks. Base yield grows. Composition changes; the modeled net APY stays in a similar range. By month 12 the depositor is being paid by real borrower interest, not by founder treasury.

What we're asking curators to believe

Three things, all independently verifiable:

1. The math works. The AdaptiveCurveIRM is Morpho's public IP. The drift formula is in their codebase. Plug in 99% util and you get the same +5/year drift speed we did.
2. The founder will execute the match-borrow. Verifiable via on-chain monitoring of the vault Deposit events and the founder wallet's Borrow events. We commit publicly on the strategy page; we report monthly.
3. The destination is real. At 90% util the IRM's drift term is zero. If we hit rateAtTarget = 9.91% during bootstrap and then ratchet util down to 90%, the rate stays at 9.91%. From there the math gives 8.03% net to depositors after the 10% performance fee.

The risks, stated honestly

LP exit buffer. At 99% util the vault has 1% liquid headroom. If a large LP wants out tomorrow they may have to wait for a borrower repayment or for their share to be matched by new deposits. That's an uncomfortable trade-off and we acknowledge it. The mitigation: during bootstrap we're largely the only depositor, so the constraint is largely self-imposed. The moment a meaningful external LP deposits, we explicitly commit to ratcheting the cap back to 85% before that deposit is matched. The cap is a configurable knob; the strategy is bidirectional.

Interest cost on the borrower. Sustained high util means the founder is paying rapidly increasing interest. We modeled this. At equilibrium (~$200K TVL, ~9.91% APR on the borrow side) annual interest is ~$19.8K against a borrower who is being rehabilitated out of pre-foreclosure. The number works because the alternative is foreclosure, which is much more expensive.

Collateral capacity. The genesis lien is capped at $275K by the signed SEA. We have roughly $240K of headroom against that cap. If external LP deposits exceed that, we need borrower #2 onboarded before match-borrow can continue. That's a manageable engineering problem — the protocol is multi-borrower by design — but it's a real ceiling on this specific bootstrap.

Reward source is finite. The Merkl reward layer is funded from the corp Privy wallet ($902 currently available, $500 committed to month 1). It's a decaying subsidy, not perpetual emissions. If we can't scale the reward budget to keep headline APY attractive while base APY ramps, the bootstrap stalls.

What's next

The timeline:

• Months 0–3 — Bootstrap. Cap at 99%. Founder is sole borrower. Merkl rewards subsidize the first LP wave. Daily monitoring of rateAtTarget.
• Months 3–7 — Ramp. External LPs arrive; match-borrow within 24h. cap tightens dynamically to keep util ≥ 95%. rateAtTarget moves through the 1–5% range.
• ~Month 8 — Lock-in event. When rateAtTarget crosses 9% APR sustained for 7 days, we stop match-borrowing. Util dilutes naturally to 90%. Webapp cap ratchets 99% → 90%. Drift → 0. Rate freezes.
• Months 8+ — Equilibrium. 8%+ net APY at 90% util. 10% LP exit buffer. Reward subsidy winds down. Genesis loan has fully rehabilitated at SPS. Onboard borrower #2 for further TVL growth.

We'll publish monthly match-borrow compliance reports and a weekly rateAtTarget snapshot starting next Monday. The live operational view of all of this is the strategy page; the source of truth for on-chain numbers is the vault itself.

References

• Live strategy dashboard — app.unblockequity.com/strategy
• Genesis vault on Morpho — 0x012f6f38…fC40E650
• DefiLlama protocol page
• Founder on Twitter — @UnblockEquity