Hi all, happy Tuesday!

Markets largely ignored Middle East tensions, and crypto rallied broadly on 4/20, with bitcoin up 2.61%, while DeFi lending markets are experiencing extreme yield spikes amid continued fallout from the KelpDAO/LayerZero exploit.

Today, we also share a few thoughts on the recent DeFi woes and the broader questions they may be surfacing around how risk is underwritten in lending protocols.

Equity markets continued to shrug off the de facto Strait of Hormuz closure, on Monday, even as between 2,000 and 3,000 ships remain stranded, and dozens that have tried to leave have been turned back by either Iranian or US military forces. The ceasefire is set to expire Wednesday, and it’s still unclear whether the parties will meet in Islamabad to discuss a resolution of the conflict.

The S&P 500 closed Monday down 0.24%, with the Nasdaq (-0.31%) and European markets faring similarly. Bitcoin fully rebounded from its weekend malaise (+2.61%), notching over a percentage point higher than gold (+1.5%) on the day, accompanied by a broad-based crypto rally.

The move was quite smoothly distributed and saw all crypto sectors above water, led modestly by crypto equities at +2.6% in line with BTC.

The liquidity crunch following the KelpDAO / LayerZero exploit on Saturday continues to roil DeFi interest rate markets. Many lending markets — even those not ostensibly linked to rsETH — have seen utilization spike to nearly 100%. That temporarily traps lenders, albeit at yield on supply shooting above 30% in some cases.

These dislocations, which have persisted for hours to days, are quite unusual and a clear sign of market stress. Capital allocators are broadly reassessing their risk and the yields commensurate with them.

Among lending protocol tokens, Euler and Fluid were favored Monday, finishing the day +8.9% and +8.6% respectively. Fluid whipped up a handy tool to relieve some of the pressure on Aave ETH loopers, allowing even whales to exit starting late Sunday.

Aave was up only marginally at +0.6% after recovering from a midday swoon. Risk manager LlamaRisk published a report covering the status of seven affected markets that had substantial rsETH reserves: Arbitrum, Avalanche, Base, Ethereum Core, Ink (via the white-labeled Tydro), Linea, and Mantle.

It remains to be seen what the full extent of the damage will be, but just to give you a sense of the tumult these markets are experiencing right now, as of this morning it’s possible to lend EURC (Circle’s euro stablecoin) at 26% APY on Aave’s Ethereum Core Instance. As one Blockworks colleague quipped, “You know the world is [messed up] when EURC has yield.” (The language may or may not have been more colorful.)

— Macauley Peterson

Who Underwrites Risk?

The past few weeks have been a wake-up call for DeFi.

In 2022, a common framing during the CeFi blowups was that centralized institutions failed, while DeFi largely worked as designed. But recent events raise harder questions. How can users justify depositing into money markets when rates are often not even competitive with Treasury yields, yet principal could go to zero overnight? What protocols can users trust if even the largest lending venue has shown poor risk underwriting?

In just the past two weeks, more than $500 million has been lost across the Drift and Kelp exploits, with contagion now spilling into other protocols, Aave most notably. Much has been written about the hacks themselves, possible recovery paths, and the Aave bad debt scenarios. More interesting to me is what these events may reveal about risk management and protocol design.

Despite scaling to over $120 billion in deposits and $50 billion in outstanding loans at its peak, money markets have remained architecturally stagnant, still largely anchored to the pool-based model popularized by Compound back in 2018.

That model has been enormously successful. It bootstrapped liquidity, enabled permissionless borrowing at scale, and generally works well when the collateral set is restricted to a minimal set of high-quality assets. But it also comes with tradeoffs, some of which recent events have made harder to ignore.

The biggest tradeoff is that pooled money markets socialize risk by default.

In Aave’s case, ETH depositors were impaired despite having no direct exposure to rsETH itself. Their exposure came indirectly because their ETH had been lent out against rsETH collateral and recycled into looping strategies. That is a structural feature of the pooled model, where lenders can be implicitly exposed to risks they may never have chosen to underwrite.

The same applies for the sUSDe looping trade. Aave has won the PT-sUSDe market since it onboarded the instrument as collateral in May 2025. The reason is fairly simple in retrospect: Aave underprices risk for that trade. USDC depositors in the protocol effectively subsidize the sUSDe loop since the unified pool model forces all users into a one-size-fits-all risk profile, regardless of their risk appetite.

Credit always has an implicit element of trust. Even in an overcollateralized model, users must trust that a protocol is underwriting onboarded collateral diligently and understand the risks embedded in configured parameters. Any money market, regardless of architecture, relies on someone underwriting risk. And where there is risk, there is always some element of trust.

In this regard, what approaches to risk management and protocol design might better preserve trust, isolate risk, and grow the pie beyond crypto-native users?

Modular lending is one possible path, but not the only one. Protocols like Morpho and Kamino can better isolate risk, though the burden of trust shifts more toward users selecting curators. And that model comes with tradeoffs too. Last month, roughly 15 Morpho vaults with non-negligible exposure (>$10k) were affected following the Resolv USR exploit. At the same time, curators like Steakhouse showcased strong risk management, underscoring the value of giving users a choice.

There are also protocols taking different approaches that have, so far, demonstrated strong risk management. One example is Maple, which takes a vertically integrated approach by creating, managing, and distributing its own products, enabling it to more directly control underwriting and risk. Another is Sky, perhaps DeFi’s most battle-tested protocol. Under its current structure, Sky remains a relatively simple base protocol that sets the broader risk framework known as Atlas, while growth and yield strategies are pushed outward to independent capital allocators.

These models are very different, but they point to a broader lesson: while architecture matters, what may matter even more is that core teams remain deeply opinionated about risk. Ultimately, only battle-tested protocols built by paranoid risk managers are likely to be the ones trusted with users’ capital at scale.

— Carlos

1. Collateral Selection is Risk Management 

Sam MacPherson from Spark published a reflection arguing that “DeFi isn’t broken, incentives are,” and that the real lesson from recent blowups is that collateral selection is risk management — brands are not a substitute for underwriting. He argues that risk and revenue must be separated (as in TradFi) and positions Spark and Sky as intentionally conservative: keep the collateral set minimal, use redundant oracle design, limit looping to ETH and wstETH to avoid hidden correlations, and add simple circuit breakers like supply/borrow rate limits to reduce capital-velocity risk. The broader point is that the industry is scaling faster than its risk frameworks, and unless there is a clear separation between risk underwriters and those optimizing for revenue, the same failures will keep repeating at a larger scale.

2. rsETH Incident Report (April 20, 2026)

LlamaRisk and other Aave service providers published an rsETH incident report explaining that the April 18 exploit was caused by Kelp’s LayerZero route being configured as a 1-of-1 DVN, which let a forged message release 116,500 rsETH on Ethereum. A large portion of that rsETH was then deposited on Aave and borrowed against, prompting Aave to freeze rsETH/wrsETH markets as a defensive measure, while emphasizing that Aave itself was not hacked and the failure originated outside the protocol. It then models potential bad-debt outcomes depending on how Kelp socializes losses, with protocol exposure varying meaningfully across scenarios.

Introducing Blockworks Investor Relations, an IR platform built for onchain businesses.

The latest Blockworks offering brings together analytics, a branded investor relations site, and integrated advisory support into a single platform. The result is a more efficient way to share your story, build trust with investors, and engage a global audience from day one.

Check out our cofounder Michael Ippolito's keynote at DAS NYC launching the new IR platform.