Pendle PT-LBTC Risk Assessment by Block Analitica
Hey everyone,
Definikola here posting on behalf of Block Analitica and B Protocol.
Pendle PT Overview
Pendle is a fixed rate protocol that works by splitting a yield bearing token into two components, a yield token (YT) that receives all interest accruing from the underlying over a specified period, and a principal token (PT) that receives a fixed value of the underlying asset at the end of the period. Because PTs do not earn interest directly, they are priced at a discount via market supply and demand, with the discount and remaining time implying a yield to maturity (YTM). The market price of a PT moves inversely to the yield (e.g. the yield increasing equates to a lower price), and converges to 1:1 with the underlying as it gets closer to maturity.
Risk Evaluation
BA will soon post the fundamentals and risk factors involved in LBTC token and the Lombard protocol as a reply to this thread. These risks are equivalent for the underlying of the PTs.
Use of Pendle introduces an additional layer of technical risk from the Pendle protocol. However, we note the protocol has been audited, and has operated for many months with TVL significantly over $1 billion without any faults or exploits. The technical risk of Pendle PTs is considered very low.
The most significant additional risk factor for PTs is duration risk, which refers to the change in market price of a PT based on changes in interest rates. As an example, if expected yields (and demand for points farming) for LBTC over the PT term increase substantially, traders will bid up the price of the LBTC YT which in turn results in a decline in price for the LBTC PT. This introduces greater volatility and uncertainty into PT pricing, and also somewhat reduces liquidity as liquidating PTs will push down their price.
Considering that the underlying LBTC yield has been 0% so far, evaluating this duration risk by discounting PTs becomes difficult since all the speculation around LBTC PT and YT tokens relies on the Babylon and Lombard points farming and their future valuation.
On the other hand, for the purpose of adding a PT-LBTC-29May2025 as collateral to the Moonwell Frontier cbBTC vault on Base, historical data of previous and existing LBTC pools on Pendle can be of help.
Liquidity Assessment (LLTVs and Supply Caps)
Following the proposition to utilize the linear price discount oracle (see below) for the PT tokens to account for their market price discount (instead of fixed 1:1 feed, which would imply setting a conservative LLTV to compensate for not accounting for market price discount of PTs), we examine the LLTV options on Morpho for the PT-LBTC-29May2025/cbBTC market.
Based on historical PT yield data for LBTC, we find the implied yield of PTs topped out at just under 8.5% on Base (on Dec 4th) for the May 2025 maturity, and 6.5% on Mainnet (on Oct 12th) for the March 2025 maturity. It’s also worth noting that this implied yield is completely supported by Babylon and Lombard points farming, i.e. currently LBTC is not generating any underlying yield (this may change in the future, of course).
Based on the historical yield data and the calculation table shown above, we can see that the 86% and 91.5% LLTV pools would be very safe for all reasonably possible PT yield environments, while the 94.5% LLTV would also be safe under current yields and would not become underwater (over 100% effective LTV) in any reasonably expected yield environment.
As risk is not the only consideration for selecting LLTV, we need to also ensure that the pool is sufficiently attractive for leverage users. Considering the maturity period of ~6 months for LBTC PT, we see that the 94.5% and 96.5% pools impose slightly higher risk with the potential for the pool to become underwater if yields increase to levels of ~25%, while the 86% LLTV pool (~7x maximum leverage) may not deliver enough capital efficiency to attract leverage users.
Based on the above, although the 94.5% LLTV is regarded as safe, we recommend focusing on the 91.5% LLTV pool (~12x maximum leverage) as a more balanced option for managing the risk/return ratio in the upcoming PT-LBTC-29May2025/cbBTC market, considering the recent deployment of the Pendle and Lombard protocols on the Base network.
Although utilizing the proposed oracle mechanism below aims to minimize the need for liquidation events to happen, we believe that we should roll out the proposed market progressively as the onchain liquidity improves.
It’s worth noting that the unwinding process of levered positions using Pendle’s PT tokens as collateral depends on a single (Pendle AMM) liquidity pool consisting of Pendle SY and PT tokens. At the time of writing, there is $4.23M of liquidity in the PT-LBTC-29May2025 pool on Base, seeing constant growth since inception to this day.
In the effort of gradual rollout, we propose the supply cap for the PT-LBTC-29May2025/cbBTC market of 30 cbBTC (~$3M at current prices).
Pendle LBTC PT - Linear Discount Oracle Analysis
Pendle has recently implemented a new price oracle mechanism that allows for pricing PTs at a discount based on a fixed yield to maturity, which allows for more accurately pricing long maturity PTs instead of using fixed redemption price. Because this does not rely on AMM TWAP pricing data, the new oracle can work without increasing risk of liquidation cascades or market manipulation.
The linear discount oracle code can be reviewed here.
Pendle LBTC PT May 2025 (Base) - Linear Discount Oracle
The above charts show how the effective LTV (market price of PT vs maximum permitted borrowable value per PT) changes based on remaining time to maturity. We see that with a 20% yield to maturity (YTM) used for the oracle, the pool would not reach dangerous effective LTVs over 100% for realized PT yields up to 40%. Using a 15% YTM results in somewhat higher effective LTVs with protection against yield spikes up to 30% for long maturities of 6 months (which is the maturity period of the LBTC-PT pool in question).
Comparing with the fixed 1:1 PT pricing strategy, we can see from the figure below that using the discounted oracle offers significantly better protection against price drops caused by spikes in implied PT yield. This also offers an improved UX, as in normal market conditions the effective maximum LTV will increase rather than decrease over the course of the PT term, reducing the likelihood of users facing liquidation from interest accrual.
Given the market conditions and the data available from July 2024 for the first LBTC pool on Pendle, we find that 30% APY is an extremely conservative upper bound estimate for potential yield to maturity for the May 2025 PT pool.
Based on the above, we calculate that using a 15% yield for the oracle for PT-LBTC-29May2025 will provide an adequate margin of safety from potential spikes in implied PT yields (and drops in PT price), ensuring that the relevant Morpho pools do not reach effective LTVs approaching 100%.
Linear discount oracle deployment on Base network by Pendle team is available here.
## Parameter Summary
We recommend to onboard the PT-LBTC-29May2025 to the Moonwell Frontier cbBTC vault with the configurations listed below:
