Understanding DeFi Fixed Rate Project Mechanisms in One Article: Zero-Coupon Bonds, Yield Tokenization, and Yield Tranching

This article is sourced from Distributed Capital, authored by James Tang and Youfei.

Fixed-rate loans are the most common type of loan in traditional finance. A fixed rate allows participants to lock in a predetermined interest rate without having to bear the risk of interest rate fluctuations. Compared to traditional finance, fixed-rate projects in DeFi offer advantages in efficiency, liquidity, transparency, and accessibility. During the development of the DeFi world, attracting large amounts of capital through fixed-rate products is an essential path for the growth of DeFi.

Currently, in addition to AAVE's stable rate borrowing, there are several protocols that provide fixed-rate services to users in various forms: these include but are not limited to:

Zero-coupon bond projects (Yield Protocol, Hifi, Notional),

Yield tokenization projects (Swivel, Element Finance, Pendle), and

Yield tranching projects (Barnbridge Finance, Horizon Finance).

1. Zero-Coupon Bond Projects

In traditional finance, a zero-coupon bond is a type of bond instrument that trades at a price lower than its face value and pays the holder the principal and interest at maturity. Currently, fixed-rate projects based on zero-coupon bonds are the most basic fixed-rate projects based on collateral-locking-redeeming.

Yield

Yield can be seen as a fixed-rate protocol based on MakerDao, which issues zero-coupon bonds in the form of yToken (such as yDAI). Users can mint yfToken by depositing collateral such as ETH. Borrowers and lenders can directly achieve borrowing and lending through yfToken.

The implementation is as follows: if the borrower wants to borrow DAI, they need to collateralize ETH, and the system automatically mints yfDAI (this step will not appear on the user interface) and exchanges yfDAI for DAI to the lender. The interest rate of yfDAI will fluctuate according to market prices. The discount of yfDAI to DAI is the interest that the borrower needs to pay. The lender can deposit DAI to obtain yfDAI at a discount, and the discount portion is the yield to maturity. The lender can choose different terms when purchasing, with different yields for different terms. After maturity, yfDAI can be redeemed back to DAI at a 1:1 ratio. The process can be simplified as follows:

Borrower A

1. Deposit collateral ETH

2. Mint yfDAI

3. Lock yfDAI in a smart contract and redeem DAI based on different terms

Due to the discount, the amount of DAI exchanged will be less than yfDAI

4. After repaying DAI at maturity, redeem the collateralized ETH

Lender B

1. Deposit DAI into the smart contract to exchange for yfDAI

Due to the discount, the amount of yfDAI will be greater than DAI

2. At maturity, exchange yfDAI 1:1 for DAI

Yield can be seen as a simplified fixed-rate protocol, with a simple user interface and basic functionalities.

Project Progress: Yield Protocol updated the latest progress on Twitter on January 26, YieldSpace v2 pool can reduce gas fees

Total Locked Amount: 2.3 million

Operational Data: Twitter-7.1K; Discord-0.81K

Backing Capital: Seed Round - Paradigm Capital, Private Round - Paradigm led, with participation from Framework Ventures, Symbolic Capital Partners, CMS Holdings, Variant, and DeFi Alliance

Governance Token: None

Mainframe (Hifi)

Mainframe can be seen as an upgraded version of Yield, with a borrowing mechanism that remains largely unchanged but offers more diversified functionalities for borrowers, lenders, and LPs. The Mainframe system consists of four parts: lenders, borrowers, collateral pools, and redemption pools. The simplified implementation mechanism is as follows:

Borrower Brad:

1. Deposit collateral ETH

2. Mint yDAI

3. Sell yDAI to the lender in exchange for DAI

There is a discount here, so the received DAI will be less than yDAI

4. After repaying DAI at maturity, redeem the collateralized ETH

Lender Lucy:

1. Lend DAI to the borrower in exchange for yDAI

Due to the discount, the exchanged yDAI will be more than DAI

2. At maturity, return yDAI to the redemption pool and burn yDAI to exchange for DAI

Guarantor Grace

1. Inject funds into the collateral pool to earn profits from flash loans obtained in the pool

The difference between Hifi and Yield is that Yield's yfDAI is automatically sold by the system through the pool, while Hifi sells yDAI directly to the lender. Therefore, the problem arises: when there is insufficient demand from lenders, the transaction cannot be completed; borrowers may need to offer a discount to obtain loans, resulting in higher borrowing costs.

In addition to basic lending services, users can also act as guarantors, injecting funds into the collateral pool to earn liquidation profits. The assets collateralized by borrowers will be used for flash loans to generate profits, which will also be injected into the collateral pool. Additionally, the collateral pool supports cToken and aToken injections, providing extra income channels for users of the Compound and Aave platforms.

Project Progress: The project was originally named Mainframe and was renamed Hifi in February this year; it will continue to use the MFT token, and token migration will occur after the new token economic model is announced. It has already been publicly tested on the mainnet.

Total Locked Amount: No data

Operational Data: Twitter-33.9K; Discord-6.1K

Backing Capital: Seed Round - IOSG Ventures, AU21 Capital, Struck Capital Crypto, etc.

Governance Token: MFT

Notional

In Notional, users can provide DAI/USDC to obtain a stable interest rate; they can also use WBTC/WETH/USDC/ETH as collateral to borrow DAI/USDC at a fixed rate.

The borrowing mechanism of Notional is not significantly different from Yield and Hifi. Its highlight lies in the LP aspect. LPs can buy the collateral of lenders at a discount when the lender is underwater, helping the lender to repay early and thus reduce risk.

Project Progress: The mainnet was launched in January 2021, and tokens have not yet been issued, expected to be launched in the v2 version.

Total Locked Amount: 17 Million USD

Operational Data: Twitter-4.1K; Discord-1.4K

Backing Capital: Seed Round - Coinbase, aparafi capital, nascent, etc.

Governance Token: None

Advanced Summary: Zero-coupon bond fixed-rate lending protocols are relatively saturated, with significant homogeneity and a lack of innovation.

2. Yield Tokenization Products

Compared to the ordinary DeFi approach of earning returns by locking assets through zero-coupon bonds, yield tokenization products greatly increase capital efficiency. Taking Pendle as an example, when the price of an asset experiences significant fluctuations due to time, savers can sell the savings interest rate (i.e., yield) of such assets in advance. Buyers can hedge these yields without purchasing the underlying assets, obtaining exposure to exchange rate fluctuations in a more efficient capital manner. For Element Finance, sellers, i.e., borrowers, can gain greater flexibility and save time without sacrificing interest rates. Buyers can purchase the underlying assets at a discount, achieving two goals at once.

Swivel

Users indirectly deposit into protocols like Compound or Aave through the Swivel protocol, where Swivel's smart contracts generate zcb ERC-20 tokens (ZCB-zero coupon bond, which has now transformed from NFT in v1 to FT) and cUSDC generated by Compound (representing future yield tokens, which have also transformed from NFT in v1 to FT). It adopts a decentralized order book model to reduce slippage during low liquidity, interest rate spreads per transaction, and capital efficiency issues.

Specific implementation method:

Below is a case using 1000 USDC and a market interest rate of 5%

Fixed-rate party enters: At the beginning of the year, pays 1000 USDC in the market, receives 1000 zcUSDC + 50 USDC from the floating party

Floating-rate party enters: At the beginning of the year, pays 50 USDC in the market, receives 1000 nUSDC

Fixed-rate party settlement: At the end of the year, pays 1000 zcUSDC, receives 1000 USDC

Floating-rate party settlement: At the end of the year, pays 1000 nUSDC, receives the interest represented by 1000 nUSDC (yield settlement: total yield at year-end - 50 USDC at the beginning of the year)

Project Progress: The testnet was launched on May 9, 2021, and the mainnet is expected to launch at the end of July 2021.

Total Locked Amount: No TVL concept in order book mode

Operational Data: Twitter-3.1K; Discord-1.2K

Backing Capital: Seed Round - Multicoin and Electric led, with participation from CMS, Divergence, Defiance, DeFi Alliance, CMT, Alex Pack, Ash Egan, Stani Kulechov, etc.

Governance Token: SWIV

Pendle

Users lock deposit certificates from deposit protocols like Aave and Compound in Pendle's smart contracts to receive OT (Ownership Token representing the right to claim the user's collateral) and XYT (representing future yield tokens). Sellers can sell XYT on DEX to lock in yield and achieve fixed-rate deposits; alternatively, they can provide liquidity in liquidity pools to earn PENDL. Buyers, on the other hand, use a smaller principal to purchase the future yield rights corresponding to the XYT.

At the same time, Pendle's improvement on AMM is its significant innovation. The AMM designed by Pendle is suitable for all assets with time decay characteristics. Initially, when the pool is created, the AMM curve is similar to Uniswap's constant product curve. However, as subsequent swaps occur, the AMM curve moves at the equilibrium point and adjusts according to time decay. Specific implementation method:

Lender A

1. Locks deposit certificates (i.e., aToken) from deposit protocols like Aave and Compound in the smart contract

2. Mints OT and XYT

3(a). Can sell XYT on DEX to obtain fixed yield

3(b). Can act as LP to earn Pendle token yield

Trader/Investor B

1. Can buy XYT and arbitrage XYT due to interest rate fluctuations

Pendle is not strictly a fixed-rate protocol. The prices of XYT and Pendle tokens will fluctuate, making the lender's yield unpredictable, and the magnitude of interest rate fluctuations (i.e., the price of XYT) also determines whether traders have room for arbitrage.

Project Progress: The mainnet was launched on June 17, 2021.

Total Locked Amount: 5.2 Million USD

Operational Data: Twitter-5.4K; Discord-3.4K

Backing Capital: Seed Round - Mechanism Capital, HashKey Capital, CMS, DeFi Alliance, imToken, Crypto.com Capital, Spartan Group, etc.

Governance Token: PENDLE

Element Finance

Element Finance divides the underlying assets (i.e., ETH, BTC, and USDC) into two parts - principal token PT (Principle Token) and yield token YT (Yield Token). Suppose a lender deposits some ETH into Element and receives newly created ptETH and ytETH. Now, they can control the principal and yield as two different tokens PT and YT. If they decide to sell PT immediately but hold YT, the underlying assets exchanged for the minted PT will decrease (for example, at 0 days to maturity, 1 ETH = 0.9 ptETH), while the held YT continues to earn interest. At this point, the lender can reinvest the underlying assets obtained.

The above operations can be summarized as follows:

Lender A

1. Deposits ETH into Element, receives ptETH and ytETH.

2. Holds ytETH, earning yield from ETH.

3. Sells ptETH at a discount, obtaining ETH.

If they want higher ETH yield.

Then repeat 1, 2, 3, 4, 5

Borrower B

Buys ptETH at a discount

At maturity, exchanges ptETH 1:1 for ETH

When the lender sells PT, the buyer can exchange it for an equivalent underlying asset at maturity, earning the price difference. For the lender, selling PT immediately can increase capital efficiency; and cyclically lending PT can leverage the interest rate. However, the lender needs to bear the risk of betting on whether the discount from selling PT can cover the leveraged interest rate. Another risk for the lender is that if there are no borrowers to take over, the lender can only sell PT at a loss. During times of significant price fluctuations, the lender also needs to manually adjust the PT price to prevent arbitrage by scientists. For borrowers, they can acquire underlying assets at a discount. In summary: the lender's cyclical operations require a high level of expertise, and cyclical lending requires the lender to bear the risk of betting; cyclical lending is not necessarily the most profitable operation.

Project Progress: The testnet was launched on May 9.

Total Locked Amount: 83 Million USD (testnet)

Operational Data: Twitter-8.1K; Discord-3.6K

Backing Capital: Seed Round - A16z, Placeholder, SV Angel, A.Capital, Scalar Capital, Robot Ventures, AAVE, Balancer, etc.

Governance Token: None

Advanced Commentary: Currently, yield tokenization projects differ slightly in market capital efficiency and model design. Swivel uses an order book model to provide customization and high capital efficiency for all market participants (liquidity providers and recipients); Pendle uses a time-depreciating coupon AMM to help liquidity providers avoid theta decay; Element uses a time-appreciating zero-coupon token AMM to help liquidity providers bypass theta, overall enhancing interoperability and reducing token transfer costs.

3. Yield Tranching Products

Yield tranching products divide returns into fixed returns (fixed rate) and floating returns (float rate): fixed returns have lower yields and risks, while floating returns have higher yields and risks.

Barnbridge

BarnBridge is a cross-platform protocol that tokenizes risk by layering yield and volatility. It uses fixed yield and volatility layers to tokenize product risks, with its native token being BOND. The project plans to launch "Smart Yield Bonds," which are fixed and floating rate products secured by DeFi yields on Ethereum, and "Smart Alpha Bonds," which are derivative tools to hedge against market price fluctuations of any ERC20 token.

Barnbridge introduces four pools: Compound/AAVE/Cream/AAVE Polygon, mining through stablecoin locking. The locked interest rates are divided into two tiers - Junior and Senior. Comparatively, Junior offers higher rates, and the APR for mining Bond tokens is also higher. Thus, Barnbridge is similar to a yield aggregator for stablecoins. The following chart shows that the interest rates for USDC in the Compound pool vary across different months.

Project Progress: Locking/LP mining started in October 2020.

Total Locked Amount: 24 Million USD

Operational Data: Twitter-18.4K; Discord-5.8K

Backing Capital: Seed Round - Fourth Revolution Capital, ParaFi Capital, etc.

Governance Token: BOND

Horizon Finance

In the first version of Horizon Finance, a no-collateral model will be adopted, allowing participants to engage in yield as two roles. One is to become a floating rate participant, i.e., the interest rate seller; the other is to participate in auctions, i.e., the interest rate buyer.

Floating rate participants can deposit currency P into the pool to mint Horizon tokens with different interest rates, auctioning them within predefined auction parameters. Here, currency P can be any token that generates income streams. Participants choose a fixed yield cap (or can choose floating) and then mint tokens (for example, setting a 9% bidding cap for yUSD between blocks X and Y) for interest rate buyers to bid on. The income distribution of the fund pool will be allocated from low to high within the fixed rate space, with excess income entering the floating fund pool. Additionally, to encourage user participation in minting, Horizon's fund pool will refund all gas costs for minting tokens, and the cost of joining existing token pools is relatively low, so users need to pay gas fees themselves.

By trading on Horizon, users will be able to actively compete to outperform general return yields or take a more passive approach to obtain satisfactory fixed rates.

Its goal is to establish a platform that does not require excessive collateral, avoiding impermanent loss for participants and eliminating the risk of protocol collapse due to excessive fixed liabilities without funds, while establishing an accurate weighted average market interest rate.

Project Progress: The beta version was officially launched in March 2021, and the mainnet has not yet been launched.

Total Locked Amount: NA

Operational Data: Twitter-3.5K; Discord-3.6K

Backing Capital: Seed Round - Framework Ventures, DeFianc Capital, Mechanical Capital, Spartan Group, Alameda Research, NGC, etc.

Governance Token: None

Advanced Summary: Yield trenching projects generally require locking for both floating and fixed rate parties, resulting in relatively low overall capital utilization efficiency. In the early stages of IRS development, they are nice to have but not essential project categories.

4. Competitive Advantages and Disadvantages Assessment

So, how do we assess the advantages of these fixed-rate projects? For non-professional DeFi players, zero-coupon bond projects are clearly easier to get started with: the user interface is more user-friendly, and the borrowing rates are clearer.

While yield tokenization projects have introduced innovations on top of existing lending protocols, improving capital utilization for both borrowers and lenders and saving time costs, the mechanisms are relatively complex and have many variable factors, making yield calculations difficult; thus, they are more suitable for professional DeFi players. As for yield tranching projects, distinguishing between floating and fixed rates and balancing yields through hedging by two types of users is inherently a zero-sum game.

Overall, most fixed-rate protocols currently have low TVL, and such projects need to draw traffic from larger lending protocols, relying on mainstream DeFi protocols. In terms of competition, leading DeFi projects' floating rates are more appealing: when people are not yet aware of the differences between floating and fixed rates, they are more inclined to go where there is more traffic.

However, as the concept of fixed rates becomes more ingrained, this sector is expected to explode in both bear and the next bull markets. The reasons are as follows: because bear markets tend to stabilize, investors are more willing to earn steady happiness at fixed rates. In the next bull market, with more large institutional funds entering, the fixed-rate market is more likely to become a mainstay, favored by players seeking stable asset returns and borrowing costs.