Why fixed-rate lending is back on the agenda
Castle Labs pulled a section from its report, Market Structure and Protocol Design, to explain why fixed-rate lending is getting fresh attention across DeFi credit. The report puts active onchain loans at roughly $28.5 billion, and says almost all demand is still piled into floating-rate lending. That is fine when markets are calm. But when stress hits, borrowing costs can spike as utilization curves shift.
When rates jump like that, some borrowers are pushed to deleverage or just get out. Bad for efficiency. In Castle Labs’ view, DeFi money markets have already cracked one problem that traditional credit still handles poorly: borrowing against collateral almost instantly. But one big hole is still there. Until a loan matures, borrowers still cannot know with certainty what their final debt cost will be.
That is why builders keep moving toward fixed-rate, fixed-term credit. Simple idea. Lenders know the yield they will earn on deposits, and borrowers know what they owe.
The report splits demand into three broad borrower and lender types:
- Duration-matching borrowers, including funds, treasuries, real-world asset issuers, and basis or carry desks that need debt maturities lined up with asset duration, redemption windows, or strategy cycles.
- Certainty-seeking borrowers, such as looping users, leveraged yield users, and traders that may not care about an exact maturity date but do need stable funding costs so spreads do not get squeezed.
- Lenders and curators, including vaults, market makers, and allocators that want to pick tenor, collateral, and return instead of taking whatever utilization spits out.
Castle Labs says earlier fixed-rate lending designs kept slamming into three problems. First, liquidity got chopped up because markets were separated by maturity, rate, collateral, and term. Second, getting out before maturity was hard unless there was already a secondary buyer, a redemption route, or some other source of liquidity. And third, the cold-start problem: lenders did not want capital stuck with no yield while waiting for the other side of a trade to show up.
The article argues that demand for fixed-term markets is climbing as more institutional capital moves onchain and as strategies like looping become more common. It also says Morpho, Jupiter, and Kamino, three protocols better known for floating-rate lending, have recently stepped into fixed-rate and fixed-term products. The report says they account for $6.83 billion in active loans combined.
Morpho Midnight and the attempt to reduce idle capital
Morpho, a long-time name in floating-rate lending, launched Morpho Midnight in July. Castle Labs describes Midnight as an intent-based zero-coupon bond lending protocol. Lenders and borrowers state intentions first. Positions are then represented through debt units, each tied to the obligation to repay one loan token by maturity, and credit units, which represent a claim on repaid loan tokens.
The core idea, as the article tells it, is to make loans tradable. That should make term handling more flexible while also giving institutions a cleaner, more predictable underwriting framework. Fixed rates are set by the price of fixed-term credit units and debt units traded by borrowers and lenders.
In Midnight, users post quotes instead of locking capital upfront. Those quotes show lending or borrowing intent in a specific market, with a set price, maturity, and collateral structure. Funds are only called at settlement through a callback process. Castle Labs says this goes straight at the cold-start problem because lenders deploy capital only after a match is executed, which improves capital efficiency.
Morpho’s team said: "By letting users earn floating rates on protocols like Morpho Blue, you can remove the opportunity cost that usually comes with waiting for quotes to be matched, and create stronger incentives to quote, increasing the total liquidity available to users."
The report points to fragmentation as another structural headache in fixed-rate markets. Every maturity, collateral type, and rate band can turn into its own venue. Midnight’s answer is to avoid tying up capital during the intent stage, so the same liquidity can be quoted across several markets at the same time. As the article puts it, “because the same capital can quote across multiple markets simultaneously, the total liquidity a single market maker can offer users equals available capital multiplied by the number of markets.”
Since going live in July 2026, active loans on Midnight have reached $3 million. Still small, the article says. But the team expects that to change because Midnight can tap Morpho’s existing network effects and wider ecosystem. Castle Labs notes that Morpho vaults now hold more than $4 billion. Once a vault adapter goes live, that capital could start quoting on Morpho Midnight and help build deeper liquidity.
One of the sharper points in the report is how Midnight handles exits before maturity. In older or thinner fixed-term markets, borrowers and lenders often had few choices before maturity. Midnight tries to improve that by making positions fungible. Lenders can sell credit units. Borrowers can buy debt units to reduce what they still owe.
Tenor Finance as an access layer on top of Midnight
Castle Labs frames Morpho Midnight as base-layer infrastructure for fixed-rate lending, then points to Tenor Finance as an access layer already being built on top of it. The article says DeFi Frontier has called Tenor “Midnight’s HIP-3.”
According to the report, Tenor keeps Midnight’s core functionality and adds several features on top:
- Auto-roll and fallback options: Tenor adds automated position rollover to cut the risk of liquidation after maturity. Independent keepers roll loans into a new fixed-rate tenor before expiry. If no fixed-rate match is found, the system can fall back to a floating-rate pool on Morpho Blue.
- Onchain OTC workflow: Users can request quotes and broadcast custom over-the-counter offers. Those offers can be shared with whitelisted counterparties for direct negotiation.
- Organization tools and access control: Tenor gives institutions role-based organization accounts that can be used to launch custom credit markets with restricted access, including borrower or lender limits tied to compliance or KYC requirements.
The report says these additions ease maturity friction. If matching liquidity is there, or fallback conditions are met, fixed-term positions can be extended with less manual work. And with the platform’s custom setup options, Castle Labs says the design fits institutional use better. The team thinks the platform could eventually serve a user base with asset managers on one side and corporates on the other.
Jupiter Offerbook and short-duration lending without price-based liquidation
Jupiter Exchange pushed Jupiter Offerbook into public beta in June 2026, around the same time Morpho Midnight’s white paper came out. The article notes that Jupiter Lend, launched last year, was the company’s first floating-rate lending product. Offerbook is its move into fixed-term credit.
Offerbook is also intent-based, but Castle Labs highlights a different feature here: there is no price-based liquidation, and the system supports fixed-term lending against long-tail assets. Loan durations are usually short, typically between 1 and 30 days. If a borrower does not repay at maturity, the lender takes the collateral directly. No liquidation process happens.
That opens the door to any long-tail collateral a lender is willing to underwrite, including NFTs, RWAs, and other assets without active price discovery. Castle Labs describes this as a distinct model because it swaps continuous price-based liquidation for maturity-based collateral transfer and creates dedicated venues for assets that are otherwise hard to support.
On Offerbook, users post intents to lend or borrow, and those intents show up in the application. Liquidity is matched when a quote gets accepted. Because users only accept once matching occurs, they can put their capital to work elsewhere before execution. The report says that setup also tackles the cold-start issue, since both lenders and borrowers can keep earning yield until a counterparty with acceptable terms appears.
Since launch, active loans on Jupiter Offerbook have reached $450,000. Even so, Castle Labs says the model’s market potential may be tougher to prove because scalability depends on lenders being willing to directly underwrite the collateral posted to the platform.
Kamino embeds fixed-rate reserves inside an existing lending stack
Kamino recently released a white paper for its fixed-rate lending protocol. Instead of building a separate fixed-rate venue, it adds fixed-rate reserve pools inside Kamino Lend.
Castle Labs says the upside of that structure is distribution. Borrowers can see a clear term structure, while lenders can quote a specific rate and tenor without fully leaving the floating-rate system. So fixed-rate lending becomes an added function inside an established market, not a separate venue that has to build everything from zero.
Each reserve pool on Kamino is defined by a rate and a tenor, for example borrowing USDC across different rates and durations. Together, those combinations form a grid.
That grid lets borrowers and lenders show where they want to trade across two dimensions: price and time. Borrowers post borrowing intents with collateral, size, maximum rate, and tenor. Lenders post conditional liquidity with the rates, durations, and amounts they are willing to offer. The grid becomes the execution layer, letting borrowers draw from available fixed-rate liquidity within predefined combinations of rate and term.
Rather than matching one by one, lenders quote on a structured matrix of predefined rates and maturities, such as 1 month at 4.5% or 3 months at 5%, creating a visible term structure and yield curve for different assets. Castle Labs says Kamino’s existing infrastructure lets borrowers either wait for liquidity through intents or directly take fixed-rate liquidity already available from the grid.
If liquidity conditions allow, Kamino can also roll loans into the next tenor automatically, in a setup the article compares with Tenor. If fixed-rate liquidity is not available, the protocol can fall back to floating rates. The goal is to reduce maturity friction and cut the need for borrowers to manage every rollover by hand.
Exits for lenders go through a withdrawal queue. If liquidity has already been deployed and an immediate withdrawal is not possible, lenders enter a first-in, first-out queue and are repaid as loans in that reserve mature. In practical terms, the article says, the maximum waiting time is capped by the reserve’s tenor.
Castle Labs also says capital is not left idle while waiting for a match. It keeps earning yield in floating-rate reserve pools, which is part of Kamino’s answer to the cold-start problem.
Fixed rates do not remove risk, but they make funding costs explicit
Castle Labs closes with a blunt point: fixed-rate lending does not erase any of the risks exposed by years of floating-rate DeFi credit. What it changes is visibility. Debt costs become explicit. And that, the article says, is what DeFi credit has been missing for a long time.
Floating-rate pools are still powerful because they make borrowing available almost instantly, but they compress all pricing into a utilization curve. Fixed-rate markets separate pricing by tenor instead. Borrowers pay for duration, lenders choose maturity and collateral risk, curators allocate across the curve, and applications can package those inputs into more predictable credit products.
The article also points to early signs of that shift elsewhere. Aave launched Stable Vaults in July, which Castle Labs presents as another early version of more predictable credit products.
That matters, the report argues, because DeFi lending is expanding into looping, basis strategies, treasury management, RWA-linked assets, and consumer-facing applications. Those users do not just need liquidity. They need clear, fixed financing terms.
Castle Labs expects competition in this segment to heat up, with more new approaches aimed at scaling fixed-rate lending. Adoption is still fairly low for now, and floating-rate lending still makes up most of the market. Even so, the stated goal is to expand the total market by serving use cases that current DeFi lending does not handle well.
The report adds that these newer products are trying to solve the same issues that held back earlier fixed-rate protocols while benefiting from stronger distribution because mature floating-rate products already sit beside them. Capital parked in floating-rate markets, for instance, can quote into fixed-rate products while still earning yield and keeping efficiency intact.
As these systems mature, Castle Labs says the lending market could get strategies that were previously hard to implement onchain, along with new growth loops for credit.
