What is Pendle?

Product introduction

Pendle is a DeFi protocol designed for the fixed income/interest rate swap market. Interest rate swaps are a very big field in traditional finance, which refers to the swap of fixed income and floating income of the same principal: for example For the same dollar-standard income, Little Y sells fixed income in exchange for current income, and Little X sells current income in exchange for fixed income.

Pendle implements something similar on the chain, first splitting out two types of income:

• Decompose the interest-earning asset (that is, the underlying asset that provides a rate of return) into YT and PT. For the same interest-earning asset, Pendle will decompose it into YT and PT that are independent of each other based on different maturities.
• YT represents the income rights of interest-earning assets. Holding YT will continue to obtain floating income from interest-earning assets until maturity, so holding YT represents obtaining floating income; PT represents the residual value of interest-earning assets after stripping the income rights. During the holding period There is no income, but after maturity, it can be exchanged for the corresponding underlying assets at 1:1. Holding PT means obtaining fixed income.

Secondly, the AMM is designed so that these two types of returns (PT and YT) can be swapped:

• Pendle packages interest-earning assets as SY and standardizes three common types of interest-earning assets (Rebase, Accumulate, and Distribute) into one type of Token. Then an AMM was designed specifically for transactions between SY and PT. Through a mechanism called FlashSwap, YT transactions can also be conducted through this AMM without the need for additional secondary markets.
• PTs with different maturities will have independent PT-SY liquidity pools. Because PTs face different duration risks, they will have different prices, which means different implied yields under different maturities.

In terms of token value capture, Pendle follows Curve and uses the veToken design. Pendle token holders can pledge tokens to obtain vePendle. The longer the pledge, the greater the amount of vePendle. By holding vePendle, you can get a share of the protocol’s income, vote on the distribution of weekly liquidity incentives, and enhance your own LP returns.

Glossary

• Base Asset: Base Asset, the principal of interest-bearing assets, no yield, such as ETH or DAI
• Interest-bearing assets: Interest-Bearing Assets, also called underlying assets, are assets with a return rate obtained after investing the underlying assets in other DeFi protocols, such as stETH to ETH and sDAI to DAI.
  a. Rebase class: Holding Rebase class Token, the number of Tokens will automatically change, and the income will be reflected by the change in the number of Tokens. The exchange rate between interest-earning asset Token and basic asset Token is always 1:1, such as stETH and aToken.
  b. Accumulate class: If you hold the accumulate class Token, the number of Tokens will not change, and the income will be reflected by the increase in the intrinsic value of the Token. The exchange rate of the interest-earning asset Token and the basic asset Token will increase with the accumulation of income, such as wstETH and cToken.
  c. Distribute class: Holding Distribute class Token, the number of Tokens will not change, and the income will be reflected through additional distribution, requiring users to manually claim, such as the liquidity incentive part of GLP and LP Token.

Team Profile

TN, the co-founder of the Pendle team, came into contact with Crypto when he was a student in 2014. After graduation, he joined Kyber Network as a member of the Funding Team and served as the head of BD. After leaving Kyber, I started some other unsuccessful ventures before starting to build Pendle. Another Co-Founder is also the author of the white paper, Vu Gaba Vineb, who previously served as Tech lead at Digix.

In an interview in May, TN revealed that its team size is about 20 people, and the team has an 8+8+4 structure, with 8 people in growth, 8 in R&D, and 4 in product design.

The Pendle team also has a high degree of transparency. Many Pendle core personnel are also tweeting doxx. In the DeFi industry where anonymous construction is the mainstay, this behavior increases user confidence. Twitter’s real-name team members include agency head Ken Chia, who was previously the Asia-Pacific head of Abra, and before that worked at J.P. Morgan; Dan, the head of growth; Anton Buenavista (three, three), the head of ecology, who previously served as a senior engineer at Kyber; Engineering Lead Long Vuong Hoang is also a Paradigm Fellow.

Financing situation

Pendle conducted a seed round of financing and IDO in April 2021. The seed transfer was about 10% and 3.7M USD was raised. The tokens began to be unlocked 3 months after the IDO and were fully unlocked one year later. Now Pendle’s early investors have Got about a 10x return :)

After Pendle became popular this year, Binxin Venture announced its investment in Pendle in the form of OTC. After the token was listed on Binance, Binance labs also announced its investment in Pendle. Recently (11.9), Spartan Group, an early supporter of Pendle, made additional investments in the form of OTC, demonstrating their confidence and recognition of Pendle.

Pendle’s initial financing background is bleak compared to competing products such as Element Finance (which received investment from a16z in the seed round, with a total financing amount of 36.4m), but it still raised enough money to help the team weather the bulls and bears.

Users & Scenarios

Pendle is a product that has the following types of users:

• PT Buyer:
  • The logic of PT is similar to zero-coupon bonds. Users who want to obtain fixed income can purchase PT directly and wait for Maturity to obtain an equal amount of underlying assets. When purchasing PT, the discount of PT relative to the underlying assets represents the fixed income under this period. rate of return. It is suitable for ordinary users who need simple and stable financial products, as well as institutional users who want to reduce risks or build more stable Market Neutral strategies.
• PT/YT traders:
  • The prices of PT and YT represent the implied interest rates under different duration risks. But markets are always divided, and traders with different views on interest rates can express their views by trading PT/YT while profiting from the trade. There is a risk of loss of principal in trading PT/YT or holding YT compared to holding PT until Maturity. The essence of trading PT is the same as trading YT, but trading YT will have greater leverage. With smaller funds, you can leverage dozens of times the implicit income of the underlying assets and expand its income exposure several times.
• Liquidity Provider:
  • Due to the characteristics of PT - there is a correlation between the price and the underlying asset, it fluctuates within a certain negative premium range of the underlying asset price. PT’s post-Maturity price is equal to the underlying asset, so the impermanent loss to provide liquidity for PT-SY is smaller. The Pendle team posted some IL backtests of the liquidity pool on Twitter: In the worst case, the IL is 0.85%, which is still very low. This type of liquidity pool with low impermanent losses has always been the favorite target of liquidity providers in the DeFi world. They use currency standards to measure profits and losses, are willing to accept floating rates of return, and are willing to use the protocol relatively deeply.

• Liquidity purchaser:
  • Liquidity purchasers refer to various DeFi protocols that need to attract funds, such as LSD protocols, RWA protocols, etc. Because Pendle is built on the interest-bearing assets of other protocols and has a token design similar to Curve. Therefore, for these DeFi protocols, incentivizing liquidity on Pendle is a more efficient way to attract funds, and also adds fixed income utility to their interest-earning assets.
• Other DeFi protocol “matryoshka dolls”:
  • Because of Pendle’s token design and the unique characteristics of the two assets PT and YT, a number of DeFi protocols based on Pendle or introducing Pendle assets to expand their own scenarios have appeared on the market. For example, StakeDAO, Penpie, Dolomite, Stella, Teller, Archi, etc.

Combining the strengths of hundreds of schools of thought – Pendle’s product design

Abstracting a certain scenario into a mathematical problem and choosing the appropriate curve (formula) to solve the problem is the key to protocol design.Most interest rate swap agreements are further innovative based on the curves of Yield Protocol, Balancer, and Notional, and have settings similar to Pendle’s dismantling of interest-earning assets into zero-coupon bonds (PT) and floating income rights (YT). .

But of course, their naming and design will be slightly different in different protocols. For ease of reading, PT and YT will be used to refer to the two types of Tokens below. We will compare the differences in protocol design between Pendle and other competing products, and analyze how to make a good DeFi product design.

Note: Other DeFi products mentioned below include: Element, Tempus, AP Wine, Sense, Swivel

Asset characteristics of PT/YT (for various protocols)

For fixed income/interest rate swap assets, there are the following 4 characteristics:

• There is a correlation between the price of PT and the underlying asset, and it fluctuates within a certain negative premium range of the underlying asset price.
• always haveInterest-earning assets decompose into the relationship PT + YT
• As Maturity approaches, the price of PT will get closer and closer to the underlying asset, and the fluctuation will be smaller. Ultimately, PT = underlying asset
• As Maturity approaches, the price of YT will fluctuate less and less. Eventually, YT loses the ability to continue to earn profits, and the price is fixed at a fixed value. In Pendle’s case, the price of YT will eventually reach 0.

Key elements of the agreement

Different protocols use different designs to meet the characteristics of asset transactions in this scenario. There are three key elements:

• Design of PT and YT
• Ways to trade YT/PT
• AMM curve design

Token design of PT and YT

The design of PT in different protocols is basically the same. PT is stripped from interest-earning assets, and PT Maturity can be redeemed as the underlying asset at a ratio of 1:1 (for example, 1PT-stETH can be redeemed as stETH worth 1ETH).

The main difference is the YT design. YT represents the right to obtain floating income from interest-earning assets. The design is generally divided into two types:

Collect YT:

• Collect YT will continue to distribute floating income from interest-earning assets to YT holders until Maturity. After Maturity, income will stop being distributed, and holding YT will become worthless.
• The price of YT at the time of transaction only represents the pricing of the future income of the interest-earning asset. After Maturity, because YT cannot obtain income in the future, its reasonable price should be 0.
• Pendle v1/v2, Swivel, and Sense use this design.

Drag YT：

• Drag YT will continue to accumulate the floating income of interest-earning assets until it reaches Maturity, which can be withdrawn in one go, similar to PT.
• Therefore, the value of YT = the value of past accumulated earnings + the value of future earnings. Before Maturity, because the value of future earnings cannot be determined, the price of YT fluctuates. After Maturity, the value of YT will be a fixed value.
  • Element, AP Wine, and Tempus use this design.

Comparing the two designs, Drag YT’s design is more conventional, easier to be accessed by other protocols, and can also be traded using ordinary AMMs.

The design of Collect YT is more ingenious:

• From a user perspective, it is naturally a waste if the realized returns remain in YT and cannot be withdrawn for further investment. From a transaction perspective, if YT pricing mixes past returns with future expectations, it will be more complicated to calculate a reasonable price.
• From the perspective of protocol design, the final price of YT will be equal to a fixed value, which is a characteristic of YT, and the price of YT will eventually be equal to 0, which makes it easier to design the corresponding AMM. This is the design idea of ​​Pendle v1 AMM.

With the protocol designed by Drag YT, the AMM design related to YT is a bit ugly. YT is ultimately equal to a fixed value. This fixed value is the accumulated income in the past, but what is the amount of accumulated income in the past? Different assets and different pools are different, and there is no universal value. Therefore, it is necessary to introduce a contract that records the accumulation of income, and then design the AMM based on this dynamic value.

Ways to trade YT/PT

Transactions can be designed in two ways:

• Double pool:
  • Separate liquidity pools need to be established for PT and YT. Generally, protocols will design a special AMM for one of the Tokens, while the other Token uses a general AMM for transactions.
  • This design is more conventional and easy to think of, but its shortcomings are obvious. An interest-bearing asset requires the establishment of two liquidity pools. The entire system has greater transaction friction, LP continues to be arbitraged, and doubles the cost of liquidity incentives…
  • Pendle v1, Element, and AP Wine all use this design. Although Tempus trades PT/YT in the same pool, if the two assets can only trade with each other and without external asset group LP, they still cannot be truly priced. and transactions, so it is essentially a dual-pool design.
• single pool
  • Use the same liquidity pool to meet the trading needs of PT and YT at the same time.This design is called FlashSwap, using the relationship of interest-earning assets = PT + YT, the Token is virtualized during the Swap process, allowing the dual-currency pool to trade three Tokens.
  • Take the process of selling 10 YT to the LP pool of SY-PT as an example: first virtualize 10 PT, synthesize it with 10 YT into 10 SY, and then swap 9 SY into 10 PT and destroy it (assuming that PT=0.9 SY at this time), The remaining 1 SY is the result of selling 10 YT.
  • Compared with the dual-pool design, the single-pool design will be more ingenious, and the benefits are obvious, as it solves the transaction friction and other problems caused by the dual-pool. Pendle v2, Swivel, and Sense all adopt this design.

AMM curve design

The design of AMM is the top priority of protocol design. AMM design needs to conform to the transaction characteristics of the asset. PT and YT assets have three characteristics:

• Interest rates always fluctuate within a certain range:For example, wstETH’s demand rate may fluctuate between 3% and 5%, and GLP’s yield may fluctuate between 5% and 20%. Therefore, AMM needs to concentrate liquidity in this range.
• As Maturity approaches, the fluctuations become smaller and smaller:Because the maturity date is getting closer and closer, the duration risk is getting lower and lower. Therefore, AMM also needs to reflect the characteristics of smaller fluctuations, that is, liquidity becomes concentrated over time.
• Approaching Maturity and the price of PT or YT will return to a constant value:PT = underlying asset, YT = 0 or the income accumulation during this period, AMM also needs to reflect this price change over time.

Most interest rate swap AMMs are innovative based on the three curves of Balancer, Yield, and Notional:

Balancer v2：

• The curve of Balancer v2 is basically the same as the constant product curve of Uniswap. The difference is that the constant product curve of Uniswap v2 can only support two assets, and the value of the two assets is 50/50, while Balancer v2 canUse parameter w to control the weight of each asset value, such as the classic 80/20 pool, and supports more than or equal to two assets.
• Pendle v1 and AP Wine have made innovations based on the Balancer v2 curve. Take Pendle v1 as an example. Pendle has designed a unique AMM for YT and SY. YT of Pendle v1 belongs to Collect YT, and the value will return to 0 after Maturity, so Pendle v1Use the time parameter t to affect the weight parameter w. When the liquidity pool is initialized, the value ratio of YT to SY is 50/50. As it approaches Maturity, YT’s w becomes smaller and its value proportion also becomes smaller: 40/60, 30/70… until Maturity: 0/100. It complies with the feature that the value returns to 0 after Collect YT Maturity.
• AP Wine also uses similar logic to transform the Balancer v2 curve. However, AP Wine designs AMM for PT and SY, so the implementation details will be slightly different from Pendle v1. This article will not go into details.

Yield Protocol：

• Yield Protocol is a fixed interest rate protocol. They use AMM to achieve fixed interest rates by selling zero-coupon bond tokens at a discount.It is consistent with the transaction scenario of PT Token. Therefore, some interest rate swap projects will directly move the Yield Protocol to their own projects unchanged for PT transactions.
• This is true for both Element and Sense. Taking Element as an example, a liquidity pool is established for the underlying assets and PT. At the beginning of the establishment of the liquidity pool, the curve is a constant product curve of x*y, and as it approaches Maturity, the curve changes to x The +y curve means that the exchange rate between PT and the underlying asset is 1:1. The curve of Yield Protocol satisfies the requirements of “PT Maturity will be worth 1:1 with the underlying asset” and “The fluctuation will become smaller and smaller near Maturity” Features.
• Sense and Element are basically the same, and there are not many changes to the Yield Protocol curve. Tempus designs AMM curves for PT and YT. The Tempus code is implemented based on Curve, but in fact the specific ideas are similar to Yield Protocol.

Notional v2：

• Notional v2 is a fixed interest rate lending protocol on the same track and with the same idea as Yield Protocol - using the principle of zero-coupon bonds to achieve fixed interest rates. However, the difference with Notional v2 is that they use a relatively flat Logic Curve, which has lower volatility when the pool is first established, which means higher capital efficiency.So the curve of Notional v2 is an improved version of Yield Protocol, we introduce the relevant formulas in detail in the appendix.
• Pendle v2 draws on the curve of Notion v2 to establish a liquidity pool for PT and SY. At the same time, it meets the requirements of “PT will be worth 1:1 with the underlying assets after maturity”, “the fluctuation will become smaller and smaller near maturity” and “ Interest rates always fluctuate within a certain range, and AMM needs to focus on three characteristics of liquidity. In our opinion, Notional v2 is indeed the better choice among these three curves. \

In addition, Swivel took a different approach and did not use AMM, but designed an order book with FlashSwap function. This choice is naturally very difficult, and there is currently no successful on-chain order book protocol in the DeFi world.

We have listed the principles and white paper links of the three curves of Balancer, Yield, and Notional in the appendix. We strongly recommend that interested readers read the original white paper to further understand the protocol design. There are many practical technical details that cannot be introduced due to space limitations. .

Summarize

DeFi’s high-rise buildings were not built in a day, nor did they come from a sudden inspiration, but after countless attempts with real money, the industry found its final path. The following table lists the design, advantages and disadvantages of each protocol in order of the time when the protocols were launched. We can see that Pendle has its own original innovations, but it also absorbed the innovations of its predecessors in v2, and finally created a product that best suits the characteristics of the scene.

*: Adopted by Pendle V2

In addition to the above protocols, interesting participants in the interest rate swap market include Flashstake, IPOR, Voltz, and Revest. These protocols target similar markets to Pendle, but their design ideas are quite different from Pendle. The starting point of this article is to compare different protocols under the same idea and summarize the protocol design methodology. Therefore, these protocols will not be introduced. It is recommended that interested readers learn about it on their own, which will also inspire new protocol design inspiration.

Pragmatic romanticism - Pendle’s operating strategy

Although the iterative improvements of Pendle V2 products compared to V1 are the foundation for the growth of the protocol, the continued growth of interest-earning assets in the DeFi field is a driving force at the macro level. Of course, Pendle has also done a very solid and beautiful job in operations, with the right time, place and people. .

Explore narratives & build collaborations

• Trial and Error mentality, finding the right narrative

As a borrower, it is crucial to find suitable interest-earning assets of sufficient size. Pendle is best known for his prominence on the LSD circuit. Looking back, it seems perfectly natural for Pendle to latch on to the LSD narrative. However, looking at the historical development process of Pendle, it has tried multiple asset classes, such as cooperation with ApeCoin and Looksrare, but these have not achieved long-term positive results. Enter LSD and the team leaderExperiment and positive feedback from the communitycommon results.

TN once mentioned in an interview: “We try many things, but we also accept that they may not necessarily work.” Especially in a bear market, this mentality is very critical. This stable mentality is probably inseparable from some major premises. One is that the interest rate market is a trillion-level market in traditional finance, but in DeFi, interest rate trading is still a very niche field. If you believe that the huge gap here will To make up for it, it is nothing more than a question of how to find a suitable path. Another point is that TN, as an “old man” who joined the industry in 2016 and has experienced cycles, has seen how AAVE and Compound transformed from unknowns to TVL’s multi-billion dollar head agreement.He also believes that the fixed interest rate market will also become an important part of the field in a few years.With such belief and dedication, one can naturally focus on more practical and pragmatic aspects such as how to find a better PMF, plan budgets appropriately, and safely survive the bear market.

Pendle is also reproducing its ability to capture important narratives, such as the RWA-related pool it has been focusing on recently. The founder of Synthetix also said:“You have to play the short-term narrative game.” Narratives in the DeFi field are changing with each passing day. How to identify the narratives worth grasping and investing in for the long term requires a lot of skills on the part of the team. But with successful experience and a good mentality, the probability of Pendle successfully capturing the narrative is worth looking forward to.

• Learn to take advantage of the situation when you are weak and provide differentiated value points in the existing interest chain

Pendle’s first major exposure in the LSD field was the launch of the Aura rETH-WETH pool in January 2023. Through joint announcements with Aura Finance, Rocket Pool, and Balancer, Pendle has successfully attracted attention beyond its own potential. At the same time, this is also a rare word-of-mouth endorsement.

So why are these established protocols willing to support Pendle? In addition to mutual conceptual recognition, the underlying reason is that Pendle also greatly helps other protocols. Pendle stacks additional APY on the basis of interest-earning assets. There is almost no opportunity cost for users to participate in this new protocol. For other protocols, the additional Pendle rewards provided by Pendle can also help them attract liquidity effectively. Currently, the Pendle War ecosystem (Equilibria, Penpie, StakeDAO) has been initially established on Pendle. Given that a unit bribe amount can often be exchanged for a higher Pendle reward, the agreement has a leverage effect by investing its growth budget in the Pendle bribery election.

Pendle is a good example of how to promote interest synergy in the DeFi ecosystem.

• Successful experiences are replicated and amplified, and the “ordinary” road to sustained growth

After the success of the Aura rETH-WETH pool, Pendle successively launched ankrETH, Stafi-rETH and other pools around the LSD ecology, basically based on Aura’s successful case. Since LSD is still in the process of continuous development, what Pendle needs to do to rely on this ecosystem is to keep up with the development of LSD and launch a new pool of potential assets in a timely manner. When it has explored a path in an ecological field, it has to do moreSolid execution amplifies existing advantages. The ideal state is that when the subjective driving force accumulates to a certain extent, the ecology can spontaneously turn around.

It’s worth mentioning that growth plays a very important role in the Pendle team. 8 people is growth in a team of around 20 people. Although the growth here may be in a broad sense (including BD, ecological construction, etc.), it is enough to show that the team attaches great importance to growth.

Expanding asset classes & public chains

• Ideal target GLP: Suitable asset characteristics & large size

For Pendle, if we want to achieve sustained growth, on the one hand, we must continue to fully understand this ecology, just like in the LSD ecology, and on the other hand, we need to find more valuable ecology.

In March 2023, Pendle officially introduced GLP (LP Token of GMX v1) and entered the Arbitrum ecosystem. GLP is a very popular asset in the Arbitrum ecosystem, consisting primarily of a basket of blue-chip cryptocurrencies while offering an ETH-based yield of 5% to 20%. In the first half of the year, the volume of GLP still maintained a certain growth at a high level. At its peak, its supply exceeded 550M. Its characteristics of interest-earning assets are very suitable for Pendle.

For Pendle, assets that have suitable characteristics, are in the spotlight and have sufficient size are undoubtedly ideal candidates. The large size also means that the depth of the pool may be sufficient to provide a better trading experience. Beyond this, the volatility of GLP’s weekly returns is attractive to swing traders, prompting more speculation and trading activity.

    https://stats.gmx.io/arbitrum

Arbitrum is currently the second largest chain on Pendle, contributing nearly 30% of TVL

    https://defillama.com/protocol/pendle

• Prioritize taking root in areas with abundant water and grass (public chain)

Chain expansion is an important source of Pendle TVL. In addition to Ethereum and Arbitrum, Pendle has recently settled on BSC and Optimism. In terms of the logic of chain selection, for Pendle, the chain itself needs to have a relatively mature interest-earning asset ecology and sufficient volume (hundreds of millions of dollars+). Because there is a conversion rate from original interest-earning assets to Pendle,Pendle’s investment makes sense only if the chain is large enough.By this standard, Pendle doesn’t have many chains to choose from.

TN once mentioned in an interview that it may enter Polygon in the future, but it is not a high priority at the moment. Why is there priority? Judging from speculation, on the one hand, it has limited energy for operations and publicity, and on the other hand, Pendle has not yet achieved the goal of listing assets completely without permission. The team needs to study the listing agreement and write the contract on its own, which takes at least a week, so the team’s energy becomes one of the bottlenecks for expansion. Pendle on BSC and OP is actually not operating well. Therefore, we need to carefully evaluate priorities, direct the team’s energy to where it can produce the most, and pick the low-hanging fruits first.

• Appropriate use of token incentives

Token incentives are a commonly used method in the DeFi ecosystem, but different project parties have different opinions on the stage at which token incentives should be implemented. For example, Kenton, the founder of Sense Protocol, which is very similar in principle to Pendle, believes that DeFi protocols should avoid token incentives before reaching PMF. Token incentives will make it difficult for DeFi products to obtain real feedback.

But judging from the example of Pendle, token incentives may lead to a certain degree of “false prosperity”, but it also helped Pendle attract a group of users and obtain product feedback earlier even when the product’s ease of use and PMF were not very good. , such as what kind of assets are suitable for Pendle’s trading form. At the same time, the TVL pushed up by token incentives is not meaningless. A high TVL will bring low slippage, making it easier to attract giant whales and institutional users, and also verify its own safety with large amounts.These are network effects for a DeFi protocol, and the costs of these attempts are borne by tokens, which will not excessively consume current cash reserves. It is still a good choice for teams that do not have sufficient funds in the early stage.

User growth & education

The current core goal of Pendle is to increase the transaction volume and TVL of the protocol. For this purpose, promoting PT is the most effective means. Firstly, the PT scenario is simple, clear and easy to promote. Secondly, the PT trading experience is indeed better. Considering the correlation between PT and YT, when the trading volume of PT increases, the trading volume of YT will also increase, killing two birds with one stone. Another aspect is to expand the audience. Given that interest rate trading is still a relatively niche market, how to lower the threshold of understanding and achieve wider acceptance is also a top priority in the next step. The user group can be simply divided into retail users and institutional customers. Different groups have different breakdown strategies.

• Pendle Earn: integrate third-party placement into as many channels as possible

In July 2023, Pendle launched Pendle Earn, replacing the previous discount asset narrative with APY. Its front-end looks like the financial products of a centralized exchange. The difference is that it is built on the chain. The interest rates and terms of the fixed income products it provides are agreed by smart contracts. It is relatively transparent and is a differentiated supplement. . Users do not need to understand complex concepts related to interest rate trading, they only need to choose products that meet their expected returns.

What Pendle needs to do is to distribute this product to as many channels as possible in an invisible or explicit form, such as integrating into CEX or various wallet providers. As a result, it may be able to reach a wider group of people and attract new liquidity.

• Provide institutional grade interest rate products

Institutions are another focus of Pendle’s expansion. To achieve large-scale adoption on the institutional side, in addition to effective BD strategies, there are also many areas that need to be improved on the product side to adapt to large capital volumes, such as a deep enough pool depth to achieve smaller impermanent losses, high capital efficiency, and achievable Diversification strategies, etc. It is worth mentioning that the interest rate structure has been formed between different stETH pools on Pendle. If the depth of such pools continues to deepen, institutions can use PT to construct their own inter-temporal “risk-free” investment strategies.

Institutional users are the main players in the interest rate market in traditional finance. Pendle also attaches great importance to this and has a dedicated institutional head (@imkenchia). In Pendle’s early days, TN would contact organizations and book 1:1 meetings to introduce them to Pendle’s revenue opportunities. This is also the entrepreneurial side of TN: willing to Get Hands Dirty,The founder is the biggest salesperson.TN actively participates in various interviews for this purpose.

At a community meeting at the end of August, Pendle disclosed that the current organization’s TVL ratio has reached 20%.

20231009 data

• Work very closely with Influencers to build trust and educate users

As Twitter is the main platform for reaching deep Web3 users, Pendle naturally did not miss it. However, although content strategy is standard for every DeFi protocol, Pendle still makes its own unique features here. Pendle’s currently officially announced ambassadors include @crypto_linn, @ViNc2453 (Chinese area), @Neoo_Nav (Vietnam). What is commendable is that each ambassador has a deep recognition of Pendle, not just a simple chat :) Take crypto linn as an example. She has expressed her appreciation for Pendle and his team in her articles many times:

https://twitter.com/crypto_linn/status/1691032585739087873

This appreciation comes from the Pendle team’s solid business capabilities on the one hand, and its thoughtful and in-depth communication style with Influencers on the other. @2lambro once mentioned that Pendle is one of the few teams that actually reads the content written by different Influencers, understands their audience, builds relationships as friends, and at the same time helps promote promotion as a professional. They schedule one-on-one meetings with Influencers to talk individually, seek feedback, and share Pendle’s future plans. Respect and sincerity will be exchanged for heartfelt recognition, True~

TN is also deeply recognized as a Founder. When building a brand for the C-side, the positive image and reputation of the founder are also a key part of it, and may even drive the product out of the industry.Building user trust with the new DeFi protocol is a lasting project.In addition to standardized strategies such as auditing, trust migration for founders/teams is also a way to go.

https://twitter.com/defi_mochi/status/1690231302577061888

Using Numbers to Observe Appearance - Interpretation of Pendle’s Current Situation

By choosing a track with potential, designing appropriate curves, and exploring assets at the forefront, Pendle’s TVL and Volume have been growing, especially this year when we are in a crypto bear market, which is even more rare:

Qualitatively, Pendle is a very good protocol, but Pendle is not without risks. Only by observing from a quantitative perspective can we judge which stage Pendle’s product development has reached, whether it has found PMF, and what the next development direction will be.

Latest data performance

Source: DefiLlama/Coingecko/Sentio/Pendle docs

• The token inflation used by Pendle for liquidity incentives can be viewed as an expense of the protocol. At present, Pendle’s daily expenditure is about 34k Pendle, and its daily income is about $1.5k. From this perspective, Pendle has no PMF and is far from profitable. However, Pendle’s fee structure has not yet been finalized. When the development of the protocol enters a mature stage, it may A new, more reasonable fee structure will be set.
• Volume represents the usage of the transaction function, and Volume/TVL can also represent whether the protocol has real demand or is just a castle in the air stacked by TVL. Pendle’s figure is 0.79%, which to some extent indicates that the currentPendle’s users are mainly liquidity providers, while the “real demand” - the proportion of PT/YT buyers is still relatively low.
• In addition to Pendle, Curve, another well-known DeFi project, is often complained about the problem of too low Volume/TVL (usually from Uni supporters lol), and Curve is actually very similar to Pendle: an AMM optimized for specific trading scenarios, in addition to trading , and the role of liquidity guidance - this role not only creates a prosperous Curve ecosystem, but also aggravates the low Volume/TVL performance, so we next compare the data of Pendle and Curve to evaluate the current status of the business. Make an assessment.

Pendle VS Curve

Source: DefiLlama/Coingecko/twitter@smyyguy

Curve’s Volume/TVL is 4.31%, which is much better than Pendle’s; daily expenditure is 448K CRV, and daily income is about $20k. Obviously Curve cannot break even. However, comparing the PS of the two, the current Pendle is 131.5 and Curve is 58.3. In addition to PS, the indicators more suitable for DeFi projects are Marketcap/TVL and Marketcap/Volume. Pendle is 0.42 and 53.0 respectively, and Curve is 0.20 and 4.73 respectively. High magnification is certain. The degree represents the market’s expectations for Pendle’s future high growth rate. This is also in line with the characteristics of the Crypto market. We encourage those who create Pattern. Innovative projects deserve higher valuations :)

Objectively speaking, Pendle is currently inferior to Curve in terms of DEX income, let alone Curve’s powerful veCRV-based ecosystem. Pendle’s current visible problems are the high slippage of YT transactions and high user learning costs. Pendle needs to increase both Volume and TVL to better generate revenue, especially the Volume needs to be increased, and Pendle’s bribery market is also in its infancy. At this stage, the amount of bribes obtained and the number of active projects cannot be compared with Curve and Balancer.

But look at the problem from the positive side:

• Pendle is actively promoting the use cases of PT Token - cooperating with lending protocols and leverage protocols, and with certain results, the flywheel effect of composability has been outlined.
• In the bear market, TVL (whether USD-based or ETH-based) and Volume are rising,A new milestone has been reached recently (let us congratulate Pendle 🎉). Both of these will bring more revenue to Pendle, but the catch is that Volume/TVL has not improved significantly. The Pendle team is about to release some new features to improve this issue.
• Compared with Curve, the TVL deposited in Pendle will generate revenue for the protocol even if there is no transaction. This is an interesting feature brought by Pendle’s unique positioning.
• There are some positive signs for Pendle’s bribery market:
  • Penpie and Equilibria are competing very actively, and there is no such thing as Convex versus Curve or Aura versus Balancer. This healthy competition prevents the value of Pendle Token from being drained by derivative protocols.
  • The Bribery Market has attracted a number of Pendle-focused players:
    • swETH has moved the bribery battlefield from Balancer to Pendle, and the ROI is very high - using a bribe of $2000 per week, 20m of TVL has been deposited in the Pendle pool, and all of this 20m will flow to the underlying assets, while Frax uses 10 A bribe that is twice that of swETH can only accumulate 40m TVL on Curve, and only 66% flows to the underlying protocol.
    • We believe that after the Shanghai upgrade, LSD can be exchanged for ETH within a few days to a few weeks, without the worry of decoupling, and the properties of interest-earning assets will be purer.The LSD bribery battlefield is also very likely to switch from Curve to Pendle.The goal of the LSD protocol will switch from pursuing price-linked stability to increasing fixed income and interest rate swap scenarios. \

There is also a reason for Curve’s low PS.The controversial behavior of Curve’s founder, high inflation rate and extremely high FDV should all be priced into the current price of Curve Token. In comparison, Pendle’s inflation rate is very restrained, and the Token has been fully unlocked.

Make Something People Want – Pendle’s future plans

To evaluate Pendle objectively, Pendle has absorbed the experience of many failed predecessors and successfully opened up a new track for DeFi. At the same time, riding on the east wind of the explosion of interest-earning assets, Pendle’s performance in protocol operations is remarkable. But on the other hand, Pendle is like a newborn baby, with no strong revenue yet, and its Volume/TVL is sluggish.

The team is also clearly aware of the current problems. In order to encourage more people to trade PT/YT Token:

• For YT: The current trading experience of YT is not good enough, and large-amount transactions have large slippages. The reason is that YT has its own leverage effect. A unit amount of YT assets usually corresponds to dozens of times the underlying assets, which requires higher liquidity depth and will cause greater price fluctuations for the same transaction volume. In this regard, Pendle has disclosed that they have plans to launch RFQ limit orders in the future. But on the other hand, with the expansion of the overall volume brought about by the increase in TVL, there is also room for improvement in YT’s trading experience.
• For PT:Another point that helps PT/YT trading is whether the Token has more uses. Lending PT as collateral is a relatively clear direction. Taking PT-GLP as an example, investors can perform revolving lending on Dolomite to achieve an annualized APY of 40-50%. Because PT’s APY is fixed, managing positions will be simpler. If there are similar cooperation with more protocols, the utility of PT will be enhanced and more people will be encouraged to purchase and use PT. This also requires the team to have strong BD capabilities. From the past, the Pendle team’s BD capabilities have always been strong. In this Twitter post, @DeFi_Made_Here summarized the 12 projects currently built on Pendle.
• Product simplification and user education: It is an indisputable fact that the threshold for understanding Pendle products is relatively high. To this end, Pendle officials have been sparing no effort. At the product level, it launched Pendle Earn, focusing on promoting PT to make it easier for users to get started. In terms of user education, it recently launched Pendle Academy (https://pendle.gitbook.io/pendle-academy/) in October to provide a detailed, one-stop trading guide. Pendle recently received an Arb Grant and will also allocate a certain amount of Arb to incentivize transactions. TN once mentioned that hot spots and narratives are fleeting, but users who understand the benefits of the interest rate trading market through education will maintain high stickiness. \

We also have some suggestions for Pendle:

• For PT: Pendle can also expand into the fixed-rate lending market on the current basis. Pendle’s curve comes from Notional, and the current flow pool is fully compatible with the needs of Notional scenarios. Therefore, you only need to add over-collateralized mint PT and liquidation functions to the protocol, and you can use the existing liquidity pool and provide liquidity for the fixed-rate lending market. This can continue to expand the use cases of PT and increase the activity of existing liquidity pools.
• Targeting the bribery market:At present, Pendle’s bribery market is not active enough, but in fact, the ROI of participating in Pendle bribery is much greater than that of Curve and Balancer. Pendle can consider using emerging assets such as swETH as benchmarks for project parties, and more actively promote new project parties to transfer bribery budgets to Pendle , as well as helping the development of Penpie/Equilibria, creating its own bribery market, and improving competition in the bribery market.

Conclusion

Pendle’s current data is relatively weak, and there are still many problems that need to be solved on the product, but Pendle is still a model of innovation in the DeFi field: innovating in a high-profile track, starting with a lower financing amount than other competitors, and after failure The second attempt is to absorb the experience of past competing products. The team is not only Crypto Nerd, but also good at BD and grasping narrative, and also willing to be practical. Do Things That Don’t Scale. Now Pendle has accumulated a very high TVL, which is much higher than previous similar protocols. It can be said that it has escaped the line of life and death. I hope Pendle can eventually verify PMF and make something people want :)

I also hope that this article will inspire everyone to build and operate DeFi innovative protocols.

appendix

Balancer v2 curve

Balancer is a spot trading AMM known for its high flexibility and customization.

The weights of different currencies can be changed by adjusting W, which is a universal version of the Uniswap v2 formula.

Whitepaper: https://balancer.fi/whitepaper.pdf

Yield Protocol Curve

Yield Protocol is a protocol that uses AMM to implement fixed-rate lending. Its AMM is specially designed to meet the needs of zero-coupon bond Token transactions.

As t approaches 0, the curve approaches the curve of x+y=k, which is in line with the characteristics of the zero-coupon bond’s exchange rate with the underlying asset after maturity is 1:1.

👉 Whitepaper: https://yield.is/YieldSpace.pdf

Yield Protocol’s white paper on AMM is definitely worth reading. It explains very well: describing the phenomenon → converting it into a formula → deriving the final formula. Although Yield Protocol has ceased operations, it has contributed to the industry a curve derivation idea that is worth learning.

Ordinary AMM is for Price, while Yield AMM is for yield, so the AMM formula is derived based on the characteristics of yield:

The collection of Fee is also derived based on yield:

The final effect of the Yield Protocol curve is: as maturity approaches, changes in the t parameter cause the shape of the curve to change, and the PT price changes, but the implicit interest rate is still continuous, and no arbitrage space is generated:

Yields cannot be negative, so half of the liquidity in the curve is actually wasted, and trading does not really occur on this half of the curve. Therefore, Yield Protocol will virtually generate the tokens required for this curve without the need for LP to provide them. The virtual Token still participates in the calculation of each transaction, so that on the basis of using the original formula, LP’s funds are not occupied unnecessarily.

Notional curve

Notion is also a protocol that uses a specially designed AMM to implement fixed-rate lending, similar to the Yield Protocol. However, the design concept of AMM is completely different. Notional’s AMM will have higher financial efficiency.

Notional chose the logit curve. By controlling the scaler parameter, the curve is relatively flat, so that Token can be traded with concentrated liquidity, and the scaler parameter is related to t. As maturity approaches, the curve becomes flatter and the Token exchange rate is close to 1. :1.

As for how to ensure the continuity of interest rates, Notional chooses to update the anchor parameter of the logit curve before each transaction to anchor the interest rate at a reasonable value and avoid arbitrage space. After re-anchoring, trade based on the scaler at that time.

👉Whitepaper：https://docs.notional.finance/traders/technical-topics/notional-amm

The Notional design idea is very novel. Yield/Curve is based on the Uniswap curve, integrating new formulas and adding new parameters to allow the Uniswap curve to deform and have new characteristics to adapt to the corresponding scenarios. With Notional, it feels like directly finding a new curve, making adjustments, and starting from scratch, which is very interesting.

However, the final effect of Notional and Yield is the same. They are both AMMs designed for “Yield”. They reflect the impact of time on the curve through the t parameter and ensure that although the curve is deforming, the interest rate is continuous.

However, Notional’s logit curve has higher capital efficiency when price fluctuations are within a certain range. Pendle v2’s white paper compares this:

👉 Whitepaper：https://github.com/pendle-finance/pendle-v2-resources/blob/main/whitepapers/V2_AMM.pdf

Disclaimer：

1. This article is reprinted from [[Buidler DAO]https://mp.weixin.qq.com/s/lpdRFSY3087ITekEZoAD8g]. All copyrights belong to the original author [@Luke @Jane]. If there are objections to this reprint, please contact the Gate Learn team, and they will handle it promptly.
2. Liability Disclaimer: The views and opinions expressed in this article are solely those of the author and do not constitute any investment advice.
3. Translations of the article into other languages are done by the Gate Learn team. Unless mentioned, copying, distributing, or plagiarizing the translated articles is prohibited.