Dion Chu

CEO of Term Labs @term_labs. Ex-bond trader at Capula & Jefferies. Federal Reserve alum. Harvard Law J.D.

[Dion Chu @dionchu]

Proud to announce our $5.5m Strategic Round led by @electriccapital with participation from @MaelstromFund, @Arete_xyz, @Deltabc_fund, @_inceptioncap, @BlizzardFund and others.

[Dion Chu @dionchu]

Vitalik’s options-based stablecoin framing is interesting because it doesn’t eliminate liquidation risk so much as move the responsibility.

Instead of an oracle/liquidation engine protecting the system in real time, P holders have to manage downside, maturity, and roll risk themselves.

If ETH tanks and the “stablecoin” holder is not quick to react or rebalance, they are the one left holding the bag.

Different failure mode. Not no failure mode.

https://t.co/De5c0YgOGk

[Dion Chu @dionchu]

Primitives are becoming table stakes.

The real moat is the complete execution stack.

More on Terminal 1 next week 👀

[Dion Chu @dionchu]

Morpho just dropped the Midnight whitepaper — a clean fixed-rate primitive.

My take: it’s the clearest sign yet that fixed-rate DeFi is moving from “build the market” to “make it actually usable.”

But here’s the key difference most people are missing…

[Morpho 🦋 @Morpho]

The Morpho Midnight Whitepaper

A noncustodial protocol for fixed rate, fixed term credit markets

https://t.co/0Sws6JiRsA

[Dion Chu @dionchu]

The hard part was never just launching a fixed-rate market.

It’s making the **full workflow** executable:

• Refinance & roll maturities
• Loop positions
• Compare routes across Aave / Morpho / Term / Pendle
• Full Safe execution + portfolio lifecycle

That’s what Terminal 1 delivers.

[Dion Chu @dionchu]

USDC borrow rates averaged 4.77% over 30 days. Intraday spikes hit 13%.

That's not a rate. That's a range.

Institutions building around financing cost need a number, not a distribution.

Weekly recap →
https://t.co/93GptUkIBW

[Dion Chu @dionchu]

Everyone's excited about RWAs coming onchain.

Here's what nobody's saying:

RWAs are fixed-rate instruments. Treasuries, bonds, credit facilities — defined terms, defined yields.

You're importing fixed-rate assets into a variable-rate lending environment.

That's not a solution. That's a mismatch.

[Dion Chu @dionchu]

Digital Assets Week starts in New York today.

One theme I’ll be watching: as more yield-bearing assets move onchain, fixed-rate rails become harder to ignore.

Borrowers do not just need liquidity against those assets. They need predictable financing costs.

If you’re in town, would be great to connect.

[Dion Chu @dionchu]

Just got back from Consensus Miami.

One theme that repeatedly surfaced across institutional, RWA, and DeFi conversations: fixed-rate infrastructure is increasingly being viewed as a necessary layer for DeFi’s next stage of maturity.

Over the past cycle, most onchain users have operated in an environment of constantly shifting rates, unstable carry, and strategies that can break simply due to utilization volatility.

As tokenized yield-bearing assets, private credit, and real-world yield products continue to grow, that model becomes increasingly difficult to scale.

Fixed-rate structures solve a different problem than simply “higher yield.”

They introduce:
• clarity around financing costs
• defined maturity outcomes
• more stable liability management
• and the beginnings of a unified benchmark curve for onchain credit markets

In many ways, this mirrors the role fixed income markets play in TradFi: creating predictability that allows larger pools of capital to actually plan around the system.

The broader takeaway from Miami was that DeFi is slowly shifting from purely maximizing yield toward building durable financial infrastructure.

[Dion Chu @dionchu]

Seeing repeat inbound for USDC borrows against yield-bearing collateral.

~$4M vs @infiniFi's siUSD is just one recent example.

Consistent theme: borrowers can't plan around variable rates.

Fixed-term demand is growing.

[Dion Chu @dionchu]

@val_usa_crypto Stay tuned!

[Dion Chu @dionchu]

https://t.co/ucNzBxQ3Wd

[Dion Chu @dionchu]

With Aave markets now effectively frozen on rsETH exposure and liquidity evaporating, we’re seeing the textbook failure mode of open-ended variable-rate pools play out in real time.

1. Shared collateral pools socialize risk across unrelated positions — one compromised LST quickly becomes bad debt that contaminates the entire market.

2. 100% utilization triggers violent rate spikes that punish legitimate borrowers exactly when they need stability most.

3. Without hard maturity dates, bad debt lingers indefinitely, distorting pricing and trapping liquidity.

Term was built around fixed-term loans and fully isolated collateral pools, designed precisely to avoid this class of contagion. Maturity enforces clean resolution: repay or liquidate, no zombie positions, no cross-position spill-over.

Worth keeping in mind for anyone actively borrowing or lending against yield-bearing assets right now.

[McKenna @Crypto_McKenna]

The rsETH incident did highlight a risk that was not overtly discussed and that is the risks associated with variable rate lending and the interest spikes that occurred from 100% pool utilisation.

This is one of the core reasons why I remain constructive on DeFi fixed-term lending where any changes in demand do not spike interest rates of fixed-term loans.

If anyone is looking to borrow at fixed-term I would highly recommend using @term_labs.

[Dion Chu @dionchu]

Great update. This is the way Term has been since Day 1. Good to see the market catching up.

[Aave @aave]

Liquidations are essential for keeping lending markets safe.

Since launch, Aave has processed over $3.3B in liquidations to protect the protocol against bad debt.

We've published a blog post introducing V4's new liquidation engine and how it improves on V3's approach.

[Dion Chu @dionchu]

Unfortunately with pre-tge tokens, recovery in the event of bad debt is all-or-nothing.

[Marius | Kamino @y2kappa]

Common questions on looping -

1/ Why would you lend stables at lower yield (5%) to someone that loops a yield bearing token with higher yield (7%)?

2/ Why not just straight up buy the 7% yielding token? (after all 7% is bigger than 5%)

3/ Why not loop it yourself 5x and make 15%? (at 80% LTV)

--

The simplest analogy(!) that answer these questions is junior v senior capital - essentially, in case of loss, loopers are liquidated first to try to make lenders whole.

Broadly, both parties have exposure to the credit risk of the yield token, however the loopers (borrowers) are the ones to get liquidated first.

Assume the yield token starts having problems and losing value (say the issuer reports an impairment of 10%) and the price immediately reflects that.

If redemptions are satisfied and liquidators can repay 1 yield token for 90% of original dollar value (so basically a 10% loss), then lenders can still be made whole.

So in reality, loopers take the first loss and both of them start suffering if impairment is bigger than Liquidation LTV buffer (1-LLTV).

--

So, you should leverage loop, buy and hold or just lend depending on your assumptions of loss and yield spread.

(broad strokes, utilization rates, slippage, liq bonuses, etc have to be taken into account)

[Dion Chu @dionchu]

@Cbb0fe Try @term_labs to lock in fixed rates

[Dion Chu @dionchu]

@PaulFrambot the backend is open, but your front end is not. Morpho decides who gets on the front page.

[Paul Frambot 🦋 @PaulFrambot]

A vault is comparable to an onchain fund, and just like traditional funds, some will perform well and others won't, but this is what we must accept and mitigate if we want to build a truly open and decentralized system.

The fact that only 1 out of ~320 vaults on the Morpho App had limited exposure to xUSD isn't evidence that the model doesn't work — quite the opposite. Morpho's isolated market + vault model meant all 319+ other vaults and their depositors, each with different risk profiles, had zero exposure.

Many assume losses equal system failure. However, in open financial systems, losses are a natural consequence of risk-taking, even when systems operate exactly as designed. As an industry, we should not fall for the fallacy that yield is risk-free. Instead, we should focus on better surfacing and educating about risks — ourselves included.

For DeFi to be the backend of finance and scale to trillions in lending volume, lending infrastructure must remain separate from risk management. We firmly believe that this open and permissionless approach is the right one, and how DeFi was intended to work.

[Dion Chu @dionchu]

Morpho is designed such that the users / vaults are concentrated to a handful of insider curators. Even if you want to choose your own markets it’s not an option on the UI. So yes, Morpho markets (lenders) are concentrated by design.

[Paul Frambot 🦋 @PaulFrambot]

Saying two isolated Morpho markets aren’t isolated just because they share a user (a vault, an aggregator, a lender…) is one of the most disingenuous — and frankly desperate — arguments I’ve seen.

By that logic, Morpho, Compound, and Aave wouldn’t be isolated either. Come on.

[Dion Chu @dionchu]

We got to do better

[Stephen | DeFi Dojo @phtevenstrong]

DeFi doesn't repeat TradFi, but it rhymes.

Black Rock valued collateral "at par" even though it had been losing actual value for years.

Now they're valueing that collateral as worthless.

This is like hard coding a strategy vault 1:1 with the dollar as it's depegging / losing NAV.