Analysis of Silo: Raised 7,500 ETH, Lending Protocol Enters 2.0 Era

BlockBeats
2021-12-13 16:04:30
Collection
What did Silo do to win first place in the ETHGlobal hackathon Chainlink Pool Prize?

Author: Lu Chang

The enormous and outdated market of on-chain lending has not seen any new developments for too long. According to DeFiLlama data, AAVE has a total locked value of $14 billion, and Compound has a total locked value of $10 billion, with these two giants operating safely and stably, while new lending protocols are mostly "AAVE like" or "Compound like." There has been little innovation in this field for a long time. However, this prolonged silence may be broken by the emergence of Silo.

Silo won first place in the 2021 "ETHGlobal" hackathon Chainlink Pool Prize.

On December 9, the non-custodial lending protocol Silo announced the completion of its Token auction on the Gnosis Auction platform, issuing a total of 100 million SILO tokens and raising 7,500 ETH, with a single SILO priced at approximately 0.000075 ETH. Regardless of where Silo's future may lead, the market's high expectations for it are undoubtedly clear.

Different Risk Exposures in "Collateralized Lending"

In the lending platform sector, attacks are not uncommon. In May 2021, the lending platform Venus was attacked, resulting in over $100 million in bad debts due to the manipulated price of the collateral asset XVS. Many users of the protocol suffered varying degrees of losses. In October, Cream Finance experienced a flash loan attack, with stolen assets amounting to approximately $130 million. The theft was caused by the attacker manipulating the price of yUSD through an oracle, creating enough borrowing limits to extract liquidity from other tokens.

Manipulating collateral prices and over-lending other more stable tokens is a common tactic seen in other lending platform attack incidents. When a lending protocol uses a risky token as collateral, all funds in the protocol are exposed to the risks associated with that token, making it difficult for even well-established projects to avoid such risks.

For example, after Cream was attacked at the end of October, AAVE faced security risks due to its support for xSUSHI, which could potentially be manipulated. Although the community subsequently disabled xSUSHI, AAVE still faced massive withdrawals. Within hours on October 30, users withdrew billions of dollars in assets from AAVE, causing the stablecoin interest rates on the AAVE platform to soar, with the DAI floating borrowing rate reaching as high as 66.65%.

In the design of most lending protocols, there exists a classic "barrel effect": the least secure collateral asset determines the overall security of the protocol. Therefore, as protocols continue to expand their list of supported collateral assets, the overall security of the protocol is continuously declining.

SushiSwap's lending product "Kashi" has somewhat addressed this issue.

Through its single vault model "BentoBox," each different lending pair can independently bear risk, and users' collateral of different tokens is not shared across different lending pools. This keeps the risk exposure of a single token contained within that vault, meaning that if a specific token experiences a security incident, it will not threaten the safety of other liquidity pools.

The barrel effect faced by well-known projects like AAVE and Compound has mysteriously disappeared on Kashi. If the price of token A becomes unpegged, only the lending pair containing A will be affected, while other pairs in the protocol can continue to operate smoothly. This allows Kashi to introduce more long-tail assets, and the lending pairs supported by Kashi are among the most numerous in leading lending projects.

However, this also brings about a new problem—liquidity scarcity. The inability to share collateral means that users' collateral cannot be utilized efficiently: you cannot collateralize A in one pool while simultaneously borrowing B and C, forcing users to pay attention to the price trends of multiple tokens—making it impossible to know which pair will face liquidation risk first.

How Does Silo Work?

In the dilemma between security and capital efficiency, how does Silo manage?

There are two important concepts in Silo's mechanism: "silo" (the literal meaning of the word) and "bridge assets."

Bridge assets and a specific token together form a silo.

(The "silo" concept is similar to the traditional LP token pair concept)

Silo's liquidity pool consists of several silos, like individual islands, ensuring the isolation of risks from different tokens. Different silos are interconnected through bridge assets (such as ETH).

(Different silos use the same bridge asset)

When a user collateralizes A to borrow B, the lending process is broken down into two steps:

  1. The user collateralizes token A in the A silo and borrows the bridge asset from the A silo.

  2. After borrowing the bridge asset, it is deposited into the B silo, using the bridge asset as collateral to borrow token B.

An interesting thing happens: although the user is essentially borrowing the bridge asset (i.e., ETH), the protocol allows the user to receive token B through the collateralization of the bridge asset.

What are the benefits of this approach? For depositors, the token B they lend is guaranteed by the bridge assets within the silo, rather than the potentially risky token A; for borrowers, they do not hold exposure to the bridge asset, and the user experience is as smooth as with other lending protocols; for the protocol, the only asset being lent out is the bridge asset from silo A.

Imagine a scenario: if token A faces a security risk and its price is artificially manipulated. A hacker deposits a large amount of token A, causing the price of the collateral token A to be severely inflated. In a traditional unified collateral model, they could simultaneously borrow loans corresponding to the inflated collateral.

However, in Silo's mechanism, regardless of how inflated the price of token A is, the maximum amount they can borrow is only the total bridge assets of silo A.

Through the reference of bridge assets and silos, it can provide users with the same high efficiency of unified collateral as previous lending protocols while isolating the risks of different tokens, ensuring that the entire protocol is not exposed to risk due to the reference of risky tokens. This design mechanism guarantees risk isolation and enhances the protocol's security without reducing capital efficiency, which has been relatively rare before.

Benefiting from this mechanism, anyone can create new markets and provide liquidity in Silo without permission, which is a notable difference from common lending protocols.

Project Overview

Silo will be governed by a DAO, with $SILO as the governance token for the protocol. Silo will adopt a governance minimization model, so the decisions the DAO can participate in are limited, mainly including the following events:

  1. Directing the assets controlled by the protocol to places that benefit protocol growth;

  2. Opening and closing the DAO's revenue mechanism;

  3. Adjusting the LTV and liquidation thresholds for each silo;

  4. Approving important product milestones.

Recently, Silo completed its seed round financing, with the specific amount undisclosed. The investors are primarily a group of angel investors in the DeFi space, including Joey Santoro from the FEI protocol, Sam Kazemian from Frax, Santiago R Santos, Ameen from Reflexer, Tyler Ward from BarnBridge, Lattice from Regan Bozman, Sherwin Lee and Keith from PSP Soteria, AiRTX, Don Ho, Quantstamp from 0xVentures, Emile from XDEFI, ShapeShift DAO, and others.

On December 9, Silo completed its Token auction on Gnosis Auction, issuing a total of 100 million SILO tokens and raising 7,500 ETH. This auction sold 10% of Silo's total supply.

The total supply of SILO tokens is 1 billion, which will be released over the next 4 years, with the specific distribution as follows:

(Token supply curve and distribution)

Genesis (10%)—distributed in a public auction, available immediately after the auction.

Community Treasury (45%)—linear release over 3 years, controlled by community governance.

Early Contributors (6.75%)—linear release over 4 years, starting 6 months after the token generation event.

Creator Contributors (21.75%)—linear release over 3 years, starting 6 months after the token generation event.

Early Community Rewards (0.2%)—airdrop to community members in January 2022.

Early Investors and Early Investment Advisors (6.30%)—linear release over 2 years, starting 6 months after the token generation event.

Future Contributors and Future Advisors (10%)—linear release over 4 years, starting 1 year after joining the DAO from the token generation event.

There are no detailed information about the Silo team; investors should be aware of the risks.

(Project team members)

ChainCatcher reminds readers to view blockchain rationally, enhance risk awareness, and be cautious of various virtual token issuances and speculations. All content on this site is solely market information or related party opinions, and does not constitute any form of investment advice. If you find sensitive information in the content, please click "Report", and we will handle it promptly.
ChainCatcher Building the Web3 world with innovators