One Article to Understand Morpho Midnight: When On-Chain Lending Meets Fixed Interest Rates and Term Markets

Author: Spinach Spinach Talks FinTech

Original Link:

Disclaimer: This article is a reprint. Readers can find more information through the original link. If the author has any objections to the reprint, please contact us, and we will modify it according to the author's requirements. Reprints are for information sharing only, do not constitute any investment advice, and do not represent Wu Shuo's views and positions.

DeFi lending has been nearly ten years, with only one main line: the floating interest rate money market.

From Aave, Compound to Morpho Blue, interest rates have always been passively discovered through utilization.

In May 2026, Morpho released the Midnight white paper. What it aims to fill is the missing piece on this main line—the fixed interest rate and fixed term.

Don’t underestimate these two words.

Fixed income (bonds, notes, credit) is an asset class with a size that surpasses the stock market globally, and its entire pricing and risk control logic—predictable funding costs, duration management, a reference yield curve—all rely on "fixed interest rates and clear terms."

On-chain lending has been doing this for many years but has always remained in the perpetual money market of floating interest rates: unable to provide the certainty institutional players need, nor to develop a proper yield curve.

This precisely is one of the structural barriers preventing institutional funds and trillions of RWA from being onboarded at scale. In other words, Midnight is not adding a new feature but is providing the underlying syntax that traditional fixed income markets lack for on-chain credit access.

This may sound like "adding an option," but the real implication is: for the first time, on-chain credit has a complete language to move from the "money market" to the "fixed income market."

I. What is Midnight

In one sentence: Midnight is a non-custodial fixed interest rate lending protocol designed for EVM.

It organizes markets around "isolated, immutable, permissionless creation, with fixed maturity," rewriting lending and borrowing into the buying and selling of a kind of "zero-coupon instrument"—lenders buy certificates, borrowers sell certificates, and both their yields and costs are embedded in the transaction price discount.

If Morpho Blue answers the question "How to make floating rate lending simple, isolated, and permissionless," then Midnight answers the next question: how to natively create a fixed interest rate, with clear maturity, and avoid liquidity fragmentation killing the market.

Following Morpho’s evolutionary mainline, we will explain the origin and development of this design.

II. From Aave to Blue to Midnight: A Clear Evolutionary Mainline

To understand the design choices of Midnight, one must first see which mainline it stands on.

First Generation: Pooling + Floating Interest Rates (Aave / Compound):

Early lending protocols emerged in an environment with thin, passive on-chain liquidity and high transaction costs. Under these constraints, aggregating all users into a single, always-enterable liquidity pool was the optimal solution to maximize liquidity concentration.

The cost was: the protocol had to make all decisions for everyone—not just settlement and bookkeeping, but also key pricing and risk parameters. This design works well when user preferences are highly homogeneous, but as assets, users, and credit scenarios diversify, and risk/liquidity/compliance preferences start to differ, a single pool cannot accommodate multiple risk profiles without fragmenting liquidity.

Second Generation: Morpho Blue—Minimal Core + Curation Layer:

Blue proposed a different architecture: markets based on isolation, immutability, and permissionless creation. The protocol itself does not judge "which assets are worth credit" or "how to allocate capital"—these decisions are deliberately left to lenders, who create and choose markets matching their needs.

In practice, most supply comes from vaults built on top of the protocol. Thus, the market layer remains thin, while curation and capital allocation form a competitive layer above the protocol.

This is Morpho’s core philosophy: keep the core as simple as possible, moving complexity to a competitive external layer.

Third Generation: Midnight—bringing fixed interest and fixed terms on-chain:

Pooling architecture and floating interest rates are inherently paired: utilization rate of the pool is regulated by an interest rate model (IRM), and interest rates are "discovered" via utilization. This mechanism is simple but has structural costs.

Midnight inherits all of Blue’s DNA—markets remain isolated, immutable, permissionless, serving as trustless primitives to build independent products and serve different jurisdictions—but replaces the interest mechanism with fixed interest rates, introducing fixed maturity and offer-based matching.

Understanding this mainline, you see that Midnight is not an entirely new species but a natural extension of Morpho’s "pushing decision-making from protocol to market/curation layer": Blue hands over rate/distribution decisions to the market, and Midnight further hands over "rate discovery" itself to market quotes.

III. Why Fixed Interest + Fixed Term?—Clarifying the Underlying Motivation

Many ask: floating interest rates are working well, why bother with fixed interest rates? Because floating rates have several unavoidable structural issues:

1. Interest rate risk is a direct obstacle for borrowers.

For borrowers needing predictable financing costs—such as institutions matching on-chain credit with off-chain fixed income liabilities—floating rates are a barrier. Funding costs fluctuate with utilization, making cash flow matching impossible.

2. Floating rates hinder the cold start of new credit scenarios.

In small markets, moderate inflows and outflows can cause utilization to fluctuate wildly, pushing interest rates to extremes. This volatility makes it hard to establish stable expectations for new markets.

3. Lenders are forced to monitor constantly.

To keep their allocations aligned with risk-reward preferences, lenders must track utilization changes and adjust positions at any time.

Fixed interest rates naturally resolve these issues.

They decouple interest from utilization: interest no longer depends on utilization but results directly from market buy/sell quotes. Borrowers get a predictable cost, lenders a predictable yield at maturity, and neither needs to chase a utilization curve.

While fixed interest rate exploration exists in DeFi (e.g., Yield Protocol), it has not become a universal base for on-chain lending—Midnight aims to do exactly that.

Fixed maturity (fixed term) is the twin premise of fixed interest. Only when a position has a clear expiry date does "borrowing/lending at a certain interest rate over a certain period" make sense; markets with different maturities form a term structure, which is the on-chain version of the yield curve.

IV. Market and Units: Rewriting Lending as "Zero-Coupon Certificate Trading"

This is the key to understanding all mechanisms of Midnight.

4.1 Market Composition

Midnight organizes markets around isolated, immutable fixed-term markets, with creation once set, unchangeable thereafter. Each market specifies:

• A loan asset (loan token);

• A maturity date;

• A set of acceptable collateral assets and their parameters (can be single or multiple collateral types).

4.2 Rewriting Borrowing and Lending with "Certificates (Units)"

Positions within the market are measured in "certificates," with a very clean logic:

• A debt certificate = obligation to repay one unit of the loan asset before maturity;

• A credit certificate = claim on those repaid assets.

Thus: buying a certificate → increasing your credit (becoming a lender); selling a certificate → increasing your debt (becoming a borrower).

Interest rates are embedded in the transaction discount. For any transaction price P > 0, the simple interest rate over the remaining period is:

r = 1 / P − 1

Example: buying a certificate at 0.95, which pays out 1 unit of the loan asset at maturity, yields approximately 5.26%. This is the same pricing logic as zero-coupon bonds or treasury bills—buy at a discount, receive face value, and the yield is in the discount.

Midnight thus fully translates "lending" into "buying and selling zero-coupon certificates," which is why it can express fixed interest rates so simply: an interest rate is fundamentally a price.

4.3 Homogenization and "Fixed Calendar Maturity": Why Liquidity Won't Fragment

This is an often-overlooked but critical design point.

Each trade involves a buyer and a seller, but the resulting position is fungible at the market level, not a bilateral relationship. Credit and debt are recorded at the market level, not tied to the specific trade that created them.

More importantly, markets mature on a fixed calendar date, not from the opening time. This means—positions opened at different times but with the same expiry are part of the same market and are fully fungible.

Why is this important?

Because in an isolated market architecture, liquidity fragmentation is the enemy: if each loan is a separate "opening date + term" instrument, even if everyone wants "90 days," the funds are sliced into countless disconnected pools.

Fixed calendar expiry cuts this problem at the root: a position entered today expiring on December 31 is the same as one entered yesterday with the same expiry, allowing them to trade and be marked-to-market with each other. Liquidity is concentrated around the "maturity date" dimension, not dispersed by opening time.

4.4 Early Exit: Four Trading Scenarios

Since credit and debt are fungible at the market level, lenders and borrowers can exit early: lenders sell certificates to reduce credit, borrowers buy certificates to reduce debt.

Rules have a clear priority—buyers will first settle their debt before accumulating credit; sellers will first settle their credit before accumulating debt.

Thus, a trade (buyer ↔ seller) falls into one of four cases depending on initial positions:

• Seller increases debt

• Seller reduces credit

• Buyer increases credit

• New debt ↔ new credit

• New credit ↔ seller reduces credit

• Buyer reduces debt

• Buyer settles debt ↔ new debt

• Buyer settles debt ↔ seller reduces credit

Early exit makes the yield curve more flexible, and because entry and exit happen within the same market, it deepens liquidity for all participants.

A detail: after expiry, trading can still occur, except that no new debt can be added (the "seller increases debt" cases are prohibited). Keeping trading after expiry allows unwinding when liquidation is unprofitable.

V. Offer Mechanism: Midnight’s True Innovation Core

If the previous section is "rewriting lending as certificate trading," then this section is "how to enable these certificates to be traded efficiently at very low capital cost." Midnight’s answer is its differentiator.

5.1 Offer: Off-Chain Quotes Without Locking Funds

Market makers (makers) express "I am willing to trade in a certain market, at a certain price, up to a certain volume" via offers. Key points:

• Offers are not broadcast on the protocol; they can be distributed via any off-chain or on-chain channel—protocol does not maintain an order book.

• Offers do not lock funds; they are simply executable intents with price and maximum size.

Takers submit offers to the Midnight contract to execute. Partial fills are allowed: any amount up to the remaining capacity of the offer. Multiple takers can fill an offer until exhausted. The contract atomically settles the referenced market—creating, transferring, or destroying the corresponding credit and debt certificates as needed.

Each offer includes a ratifier (approval contract) with embedded verification logic, invoked when the offer is filled. Usually, it verifies signatures on the maker’s public key.

This modular design allows makers to use different signature schemes (e.g., passkeys, quantum-resistant schemes) or custom verification logic—laying the groundwork for "one signature approval for multiple offers."

5.2 Maker Callback: Funds Are Only Drawn at Execution

This is the core of the mechanism.

Offers can specify a callback to be executed at the moment of trade, allowing makers to gather the necessary funds or collateral only when the offer is filled, not beforehand.

This means makers can keep their capital elsewhere earning yield until the offer is actually taken.

The white paper’s example: a lender can keep funds in a Morpho Blue market earning yield, while also posting a fixed-rate offer on Midnight. When the offer is filled, the callback pulls funds from Blue and completes settlement (assuming sufficient liquidity).

This is especially useful for rolling fixed-term exposure. Near expiry, borrowers can buy back or repay debt atomically within the same transaction, rolling their exposure; lenders can shift credit exposure from one expiry to another without withdrawing idle funds.

5.3 Multi-Market Quotes, Consumption Groups, and Merkle Roots: Covering All Markets with One Fund

Callbacks enable a stronger capability: a maker can use the same liquidity to post multiple offers across multiple markets—key to combating liquidity fragmentation.

But there’s a risk: if a single 10 ETH fund supports three offers in markets A, B, and C, can it be drained for 30 ETH?

No.

Midnight uses consumption groups:

• Multiple offers belonging to the same group share a fill budget.

• When any offer in the group is executed, it deducts from all offers’ remaining budgets.

• Once the budget is exhausted, no further fills occur within the group.

Thus, the maker’s actual exposure is limited by the group’s budget, not the sum of individual offer sizes.

White paper example: a lender has 10 ETH, posts offers 1/2/3 in markets A/B/C sharing a 10 ETH fill budget. A borrower consumes 3 ETH from market B, reducing remaining budget to 7; another borrower consumes 7 ETH from market A, depleting the budget to zero and consuming all 10 ETH. All three offers are now fully filled.

One fund, multiple quotes, controlled exposure.

To scale this efficiently, ratifiers can approve Merkle roots of offer sets: the maker signs once, and offers are posted with Merkle proofs. When filled, the proof is presented to verify inclusion.

This combines signature efficiency with capital efficiency.

Connecting sections 5.1–5.3, you see that Midnight effectively eliminates the implicit capital cost of "order book" offers in traditional exchanges: in conventional systems, conditional liquidity (e.g., "execute at a certain rate and size") requires locking funds upfront, which becomes prohibitively costly with multiple expiry dates and isolated markets. Midnight’s "offered without locking funds, executed on demand" approach enables the market to operate before stable trade flows are established—solving the cold start problem.

5.4 Routing: Off-Chain Search, Not Centralized Order Book

The protocol does not enforce a quote queue, but the routing layer naturally compares offers by price. The issue: the protocol does not guarantee any offer’s executability (considering callback success, group exhaustion, gas costs).

Therefore, a taker seeking the "best executable liquidity" faces a real search problem. This process is called routing, happening outside the protocol, and anyone can do it.

This makes Midnight fundamentally different from a centralized limit order book (CLOB):

• The protocol does not maintain a standard order queue;

• No price-time priority is enforced at the protocol level;

• No capital is reserved by the protocol.

In other words, Midnight shifts the complex "matching/routing" task outside the protocol to a fully competitive solver/router layer.

The core responsibility: take a submitted offer and atomically settle it.

5.5 Tick Grid: Using Interest Rates Instead of Prices

Midnight sets a minimum tick size—like stock price increments of one cent.

The logic: if prices are infinitely divisible, market makers will undercut each other with tiny differences, leading to a "race to the bottom" and reduced liquidity.

The clever part: the tick size is based on interest rates, not prices.

Why not divide prices equally? Because price and interest rate are not a fixed one-to-one mapping—reducing price by 1% in a one-month expiry market corresponds to a large annualized rate, while in a one-year expiry market, it’s much smaller. Equal price steps can thus correspond to very different interest rate changes depending on maturity.

Market makers think in interest rates, not prices.

Midnight’s solution: adjacent tick levels differ by a fixed interest rate percentage (default 2%), ensuring consistent interest rate jumps across maturities.

This grid can be refined: start with 2% coarse ticks, then tighten to 1% or 0.5% as market depth grows. The design ensures that finer ticks are nested within coarser ones, so existing quotes remain valid when increasing precision.

This allows the market to smoothly improve quote granularity without invalidating existing offers, similar to how stock exchanges provide smaller tick units for better liquidity.

VI. Settlement Mechanism: More Gentle on Borrowers, Fairer Loss Sharing

Fixed maturity introduces some new considerations for liquidation, which Midnight’s mechanisms address.

The overall goals: be gentler on borrowers during liquidation, and share losses more fairly. Let’s review key mechanisms—without formulas, just what they do and why.

6.1 When Are Positions Liquidated?

Borrowing capacity depends on the "discounted market value" of collateral: each collateral type is discounted by its own LLTV (liquidation loan-to-value). The sum of discounted collaterals sets the borrowing limit. If debt exceeds this, the position becomes "liquidatable."

During liquidation, a third party repays part of the debt and takes the collateral at a discount—this is the liquidation incentive.

Each collateral has its own price feed and discount rate, allowing risk separation within a market.

6.2 How Much Discount Can Liquidators Get? Adjustable per Collateral

Liquidators are motivated because they can buy collateral below market price—this discount is their reward.

Midnight’s innovation: the maximum discount is not uniform across all markets but adjustable per collateral type ("liquidation cursor" in the white paper).

Logic: smaller discounts leave more over-collateralization for the borrower, reducing bad debt risk; larger discounts attract liquidators to handle difficult collateral.

Compared to Blue, which uses a single uniform discount, Midnight offers finer control, improving risk management.

6.3 Liquidation Only to "Just Healthy"—No Full Liquidation

When a position is unhealthy, it can be liquidated, but the amount repaid is capped—liquidators can only bring the position back to "just healthy," not fully close it (this is called "restorative liquidation").

Why is this important for fixed maturity markets?

Because borrowers must always have sufficient collateral to cover the full amount due at expiry. Allowing liquidators to close out completely at the first breach would force borrowers to post full collateral upfront, even if only part of the position is overdue, penalizing early or partial defaults.

The exception: if the remaining collateral is too small to be worth liquidating, a full liquidation is permitted to avoid leaving unmanageable residuals.

6.4 Post-Maturity: Incentivizing "Gradual Price Increase" to Avoid Pitfalls

After expiry, rules tighten: if debt remains unpaid, the position can be liquidated—even if it appears healthy on paper—since the lender is entitled to repayment at maturity.

But this is often just a borrower being late, not insolvent.

Midnight does not reward late repayment immediately; instead, it gradually increases liquidation incentives over about 15 minutes, like a Dutch auction, encouraging timely repayment without allowing liquidators to exploit late borrowers excessively.

This ensures that, ultimately, someone will settle, but late borrowers are protected from excessive penalties.

6.5 Recognizing Bad Debt and Front-Running

If collateral drops too much, and liquidation cannot recover the debt, the shortfall becomes a bad debt, shared proportionally among lenders.

Blue records losses only after collateral is fully seized, which can delay recognition of insolvency, allowing informed lenders to withdraw early.

Midnight records losses immediately upon first liquidation attempt, minimizing window for front-running and unfair advantage—this promotes fairness and timely risk accounting.

VII. Access Control and Authorization: Interfaces for Compliance and Institutions

7.1 Gate: Two Types of Optional Gatekeeping

Midnight supports flexible access controls. When creating a market, it can specify up to two gate contracts (fixed thereafter). When attempting sensitive operations, the protocol calls these gates:

• Enter gate: controls who can establish or increase positions—used for KYC, whitelists, etc.

• Exit gate (not explicitly mentioned but implied): allows withdrawal or repayment.

Important: gates only control "entry," not "exit." Even if entry is blocked, participants can still withdraw, repay, or retrieve collateral.

This design ensures gates are external, upgradeable, and cannot lock funds inside the market, avoiding custodial risks.

The liquidation gate controls who can perform liquidations, e.g., only approved liquidators.

For RWA and institutional credit, these gates enable compliant markets—white-listed participants, approved liquidators—built on immutable primitives without rewriting protocols.

7.2 Delegated Authorization: Coarse-Grained, Delegatable

Midnight offers a simple, coarse authorization primitive:

• An account can authorize another address to act on its behalf within the protocol, avoiding per-operation signatures.

Common uses:

• Allowing a keeper to manage rolling positions at expiry;

• Allowing a router or bundler to perform repay, collateral withdrawal, or entering new markets atomically;

• For lenders, depositing funds into vaults that operate on their behalf.

Note: this authorization is global: once granted, the delegate has full control over the authorized account’s Midnight state—can roll positions, withdraw collateral, incur debt, or revoke permissions.

The protocol does not provide fine-grained permissions per operation or market. The authorized address must be fully trusted or governed by code constraints.

To restrict permissions further, a middle-layer contract can be used: it holds full protocol authority but exposes limited interfaces. Vaults are such middle-layer contracts—they have full control but are coded to only operate within certain markets or limits, and depositors only hold shares.

Thus, "who can do what" logic is embedded in the middle-layer contract, while Midnight recognizes only "full authority" or "no authority," aligning with Morpho’s "minimal core, complexity moved outside" philosophy.

VIII. New Fee Types: Settlement Fees and Continuous Fees

Midnight can charge at most two types of fees—settlement fees and ongoing (continuous) fees—both written into the contract with non-upgradable caps, ensuring participants know the maximum protocol can collect.

Default fee rates are set based on the loan asset but can be overridden per market.

Settlement fee: charged per trade, represented as a spread—an extra margin inserted between buyer and seller prices. The fee is paid by the taker. It is linearly scaled over the remaining period but capped so that annualized fees do not exceed 50 basis points (0.5%).

Continuous fee: accrues over time on outstanding loans, paid by the lender when they reduce or withdraw their credit. It locks in the fee rate at the time of loan creation; even if the protocol raises rates later, existing positions are unaffected. The cap is 1% annualized.

IX. What Does This Mean? Some Judgments for Practitioners

Having explained the mechanisms, the final "so what" is worth reflecting on.

I believe Midnight’s significance can be summarized as follows:

1. It completes Morpho’s vision, pushing on-chain credit from "money markets" into "fixed income markets."

Blue + vaults provided isolated, immutable floating-rate markets and a curation layer; Midnight adds the missing primitives—fixed interest and fixed maturity. Multiple expiry markets form an on-chain native term structure / yield curve.

This step enables on-chain markets to speak the language of traditional fixed income.

2. Its core abstraction is essentially bringing the microstructure of fixed income markets onto the chain.

Zero-coupon discount pricing, calendar expiry, homogeneous secondary liquidity, quote-based, off-chain distribution and routing, tick grids, expiry liquidation—these almost directly mirror traditional bonds and notes.

But Midnight builds on Morpho’s "isolation / immutable / permissionless creation" DNA, maintaining DeFi’s trustless composability while borrowing from TradFi’s mature market microstructure.

3. "Quote-based, capital-efficient trading" is an underestimated engine.

For professional market makers, this means the same capital can cover dozens of markets and multiple maturities, with exposure precisely constrained by consumption group budgets. This reduces opportunity costs of providing conditional liquidity and is the key to solving liquidity fragmentation and cold start in isolated markets.

Who can excel at off-chain routing and solver design will reap the structural benefits.

4. For RWA and institutional credit, this is almost tailor-made primitives.

Institutions need predictable costs and clear maturities—fixed interest + fixed term; RWA assets often have a term structure, and on-chain credit can finally match durations.

Gate and liquidation controls enable compliant access and designated liquidation—KYC, whitelists, permissioned markets—built on immutable primitives, with "entry only" discipline avoiding custodial risks.

5. For the Curator / Vault layer, this opens a new product space.

Just as Blue fostered a curation ecosystem, Midnight can support structured fixed-rate, fixed-maturity credit products—layered by expiry, yield curve strategies, and packaging fixed income for institutions.

Risk curation expands to include not only collateral management but also managing term structures, rollovers, and overdue liquidation pathways.

This completes the translation, preserving the original structure and content fidelity.