Yield Magnetism and System Resilience: How to Rationally View the YT Arbitrage of Ethena × Pendle?

Author: TEDAO

Introduction:

As Ethena's popularity grows, a crowded arbitrage chain is operating at full speed: collateralizing (e/s) USDe to borrow stablecoins on Aave, buying YT/PT on Pendle to generate returns, and then leveraging some of these positions back into Aave to earn external incentives like Ethena points. The result is a sharp increase in PT collateral exposure on Aave, pushing the utilization rate of major stablecoins to over 80%, making the entire system more sensitive to any disturbance.

This article will delve into the operation and exit mechanisms of this funding chain, as well as the risk management designs of Aave and Ethena. However, understanding the mechanisms is only the first step; true advancement lies in upgrading the analytical framework. We're often accustomed to using data analysis tools (like Dune) to review the "past," but what's missing is precisely how to clearly see the various possibilities of the "future" and truly implement this approach—first define risk boundaries, then discuss returns.

How Arbitrage Works: From the "Return Side" to the "System Side"

Let's first look at this arbitrage path: deposit eUSDe or sUSDe (sUSDe is staked eUSDe with native returns) in Aave, borrow stablecoins, and then buy YT/PT on Pendle. YT represents future returns, while PT, stripped of its returns, can always be purchased at a discount. Holding until maturity allows for redemption at a 1:1 ratio, profiting from the difference. Of course, the real "big money" comes from external incentives like Ethena points. The PT obtained in this way, since it can be used as collateral on Aave, becomes the perfect starting point for a revolving loan: "Pledge PT → Borrow stablecoins → Buy PT/YT → Pledge again." The goal is to leverage relatively certain returns to achieve highly flexible returns like Ethena points. How has this funding chain reshaped the lending market?

• Aave's Exposure and Second-Order Effects: USDe-backed assets have gradually become the mainstream collateral on Aave, reaching a share of approximately 43.5% at one point, directly driving up the utilization rate of the dominant stablecoins USDT/USDC.

• Crowding on the Borrower Side: Following the introduction of USDe eMode for PT collateralization, USDe borrowing reached approximately $370 million, of which approximately $220 million (≈60%) served leveraged PT strategies, causing utilization to surge from approximately 50% to approximately 80%.

• Concentration and Rehypothecation: USDe supply on Aave is highly concentrated, with the top two entities holding a combined 61%. This concentration, combined with circular leverage, amplifies returns while also exacerbating the system's fragility.

The rule here is simple: The more attractive the returns, the more crowded the cycle, and the more sensitive the entire system. Any slight fluctuation in price, interest rates, or liquidity will be ruthlessly amplified by this chain of leverage.

Note: The core on-chain data cited in this article is primarily based on a Chaos Labs report released on July 17, 2025, and related market observations. Due to the dynamic nature of on-chain data, readers are advised to consult relevant data analysis platforms for the latest information.

Why "Exit" Is Difficult: Pendle's Structural Constraints

So, how do you exit? There are two main paths to deleveraging or closing the aforementioned rotating positions:

• Market-based Exit: Sell PT/YT before maturity and redeem them for stablecoins to repay the outstanding balance.

• Hold to Maturity Exit: Hold PT until maturity, redeem it 1:1 for the underlying asset, and then repay the outstanding balance.

Why is exiting so difficult? The main difficulty stems from two structural constraints of Pendle:

• Fixed Term: PT cannot be redeemed directly before maturity and can only be sold in the secondary market. If you want to "deleverage quickly," you must rely on the performance of the secondary market, facing the dual challenges of depth and price volatility.

• AMM "Implied Yield Range": Pendle's AMM operates most efficiently within a pre-defined implied yield range. If market sentiment shifts and yield pricing falls outside this range, the AMM may become "inactive," forcing trades to be executed on a thinner order book, dramatically increasing slippage and liquidation risks. To prevent risk spillover, protocols like Aave deploy PT risk oracles: when the PT price falls to a certain floor, the market is frozen. This prevents bad debts, but also means it's difficult to sell PT in the short term, forcing you to wait for the market to recover or hold it until maturity.

Thus, while exiting during stable market conditions is generally straightforward, when the market begins to reprice and liquidity becomes congested, exiting becomes a major friction point, requiring a contingency plan to be prepared in advance.

Aave's "Brakes and Buffers": Ensuring Orderly and Controllable Deleveraging

Faced with this structural friction, how do lending protocols like Aave manage risk? They have built-in "brakes and buffers":

• Freeze and Floor Mechanism:If the PT price hits the oracle's floor price and remains there, the relevant market can be frozen until maturity. After maturity, the PT naturally decomposes into the underlying asset and can then be safely liquidated/unleashed, minimizing the liquidity misalignment caused by the fixed-term structure.

• Internalized Liquidation:In extreme cases, the liquidation reward is set to 0, forming a buffer before disposing of the collateral in stages: USDe is sold in the secondary market after liquidity has recovered, while PT is held until maturity, avoiding passive selling on the illiquid order book in the secondary market, which can amplify slippage.

• Whitelist Redemption: If a lending protocol is whitelisted on Ethena, it can bypass the secondary market and directly redeem the underlying stablecoin with USDe, reducing impact and increasing returns.

• Boundaries of Supporting Tools: When USDe liquidity is temporarily tight, Debt Swap can swap USDe-denominated debt for USDT/USDC; however, due to E-mode configuration constraints, migration has thresholds and steps, and requires sufficient margin.

Ethena's "Adaptive Foundation": Supporting Structural and Custody Separation

Lending protocols have "brakes," while the asset-backing side requires Ethena's "automatic transmission" to absorb impact.

• Regarding support structure and funding rate status: When funding rates fall or turn negative, Ethena reduces its hedging exposure and increases stablecoin support. In mid-May 2024, the proportion of stablecoins reached ~76.3% before falling back to ~50%, still relatively high compared to previous years, enabling proactive pressure relief during negative funding cycles.

• Furthermore, regarding buffer capacity: In the extreme LST slashing scenario, the net impact on USDe's overall support is estimated to be approximately 0.304%. The $60 million reserve is sufficient to absorb such a shock (accounting for only approximately 27% of it), thus limiting the actual impact on the peg and repayments.

• Asset custody and segregation are key: Ethena's assets are not held directly on exchanges, but rather through over-the-counter settlement and asset segregation via third-party custodians (such as Copper and Ceffu). This means that even if the exchange itself experiences operational or solvency issues, the ownership of the collateral assets remains independent and protected. This segregated architecture enables efficient emergency response procedures: In the event of an exchange outage, the custodian can void open positions after missing a certain number of settlement rounds, releasing collateral and enabling Ethena to quickly migrate hedged positions to other exchanges, significantly shortening the risk exposure window.

When dislocation primarily stems from "implied yield repricing" rather than a loss of USDe backing, the bad debt risk can be managed with the protection of oracle freezes and tiered disposals. The real focus is on preventing tail events involving backing losses.

What You Should Watch For: 6 Risk Signals

Now that we've covered the theory, what specific indicators should we look at? The following six signals are highly correlated with the correlation between Aave, Pendle, and Ethena and can be used as a daily dashboard for monitoring.

• USDe Borrowing and Utilization: We continuously track USDe's total borrowing volume, the proportion of leveraged PT strategies, and the utilization curve. The utilization rate has consistently remained above ~80%, indicating a significant increase in system sensitivity (from ~50% to ~80% during the reporting period).

• Aave Exposure and Second-Order Effects of Stablecoins: Pay attention to the proportion of USDe-backed assets in Aave's total collateral (e.g., ~43.5%) and its impact on the utilization of core stablecoins like USDT/USDC.

• Concentration and Rehypothecation: Monitor the deposit ratio of top addresses. When the concentration of top addresses (e.g., the top two combined) exceeds 50-60%, be wary of potential liquidity shocks caused by their aligning movements (peak value >61% during the reporting period).

• Closeness to the Implied Yield Range: Check whether the implied yield of the target PT/YT pool is close to the AMM's preset range. Closeness to or outside the range indicates decreased matching efficiency and increased exit friction.

• PT Risk Oracle Status: Monitor the distance between the PT market price and the Aave Risk Oracle's lowest price threshold; approaching the threshold is a strong signal that the leverage chain needs an orderly deceleration.

• Ethena Support Status: Regularly review Ethena's published reserve composition. Changes in the proportion of stablecoins (e.g., from ~76.3% to ~50%) reflect its adaptation strategy to funding rates and the system's buffer capacity.

Limit: Set a size cap on positions that rely on "sell before expiration."

Trigger: Implied yield exceeds the range / market depth plummets / oracle floor price approaches.

Action: Increase cash and margin ratios and adjust exit priorities; if necessary, implement a "reduce only, no increase" freeze period.

Boundary 3: Oracle Status

Price approaches the lowest price threshold or triggers a freeze, indicating that the chain has entered an orderly deceleration and deleveraging phase.

Limit: Minimum price difference (buffer) from the oracle floor price and minimum observation window.

Trigger: Price difference ≤ preset threshold / freeze signal triggered.

Action: Phased reduction of positions, enhanced liquidation alerts, implementation of debt swap/deleveraging SOPs, and increased data polling frequency.

Boundary 4: Tool Friction

Debt Swap and eMode migration are effective during periods of stress, but they come with frictions such as thresholds, waiting times, additional margin, and slippage.

Limits: Available quota/time window for the tool and maximum tolerable slippage and cost.

Trigger: Borrowing rate or waiting time exceeds threshold/trading depth falls below lower limit.

Action: Reserve capital redundancy, switch to alternative channels (gradually close positions/hold until maturity/redemption through whitelist), and suspend strategy expansion.

Conclusion and Future Directions

In summary, Ethena x Pendle arbitrage connects Aave, Pendle, and Ethena into a transmission chain from "yield magnetism" to "system resilience." The circulation of capital increases sensitivity, while structural constraints on the market raise exit barriers. Protocols, however, provide buffers through their own risk management designs.

In DeFi, the advancement of analytical capabilities is reflected in how we view and use data. We are accustomed to using data analysis tools like Dune or DeFiLlama to review the past, such as tracking position changes of top addresses or trends in protocol utilization. This is crucial, as it can help us identify systemic vulnerabilities such as high leverage and concentration. However, its limitations are also obvious: historical data presents a "static snapshot" of risk, but cannot tell us how these static risks will evolve into dynamic systemic collapse when market storms strike.

To clearly identify these potential tail risks and deduce their transmission paths, it is necessary to introduce forward-looking "stress testing"—This is exactly the role of simulation models. It allows us to parameterize all the risk signals mentioned in this article (utilization, concentration, price, etc.) and place them in a digital sandbox (a joint model of the core mechanisms of the Aave, Pendle, and Ethena protocols) and repeatedly ask, "What if...":

• If the price of ETH plummets 30% At the same time, if the funding rate turns negative, how long can I hold my position?

• How much slippage do I need to withstand to exit safely?

• What is the minimum safety margin requirement?

The answers to these questions can't be found directly from historical data, but they can be predicted in advance through simulation modeling, ultimately helping you develop a truly reliable execution playbook. To get started, consider the Python-based industry-standard cadCAD framework, or try HoloBit, a next-generation platform based on cutting-edge Generative Agent-Based Modeling (GABM) technology, which offers powerful visualization and code-free capabilities.