Working Group Proposal for FEI Lend - the integrated lending protocol native to FEI

Founding Members:
Defi3 - Lead Protocol Engineer; Vice President, US investment bank.
KHBW - Internal Audit and code support; Blockchain co. Co-founder, Protocol Designer
Taurus (Jun) - GUI and front end design; Polkadot Developer, GUI designer
Blueplanet - Protocol Engineer; Multi-year solidity experience
Frank Luo - Protocol Engineer;
Cozeno - Liaison, group manager;

One universal feedback from every section of our community is the limited use case for FEI, as well as the slow pace of integrations with other Defi applications. Coupled with calls to extend the utility for TRIBE, we therefore propose to create a lending protocol natively integrated with FEI; which will allow users to lend out and borrow FEI, TRIBE, or ETH.

After examining the field of similar applications, we have determined that COMP represents the nearest peer model of sound custodianship and safety. We have approached @leshner, and he has graciously signaled his support for a project based on a fork of COMP.

For those unfamiliar with COMP, FEI Lend will initially have the following features:

Allow users to deposit FEI, TRIBE, or ETH and receive a ERC20 Token as a certificate of deposit.
Borrow funds by using the said ERC20 Token as collateral,
The market rate for borrowing and deposits are set by a unified Demand Curve, modified for utilization ratio (ie supply and demand).
Collateral assets can be liquidated if the outstanding balance of the borrowing account exceeds the sum allowed by the collateral factor.

We will publish materials outlining every technical single parameter in the immediate future.

Governance:

We propose that FEI Lend would be wholly owned by the TRIBE DAO, and all proceeds or potential profits should be ultimately attributable to the TRIBE token. But just like Central Banks around the world do not directly participate in the commercial bank business, we propose that the assets and liabilities of FEI Lend would be segregated from the PCV. FEI Lend would operate independently of the PCV, and only remit potential profits back to the PCV as a special dividend when mandated by the DAO.

To further elaborate on the governance of FEI Lend, we propose that FEI Lend would be temporarily controlled by a multisig guardian account. We believe it is not practical at this point to create a separate DAO for FEI Lend. Any changes to the code would be submitted to the TRIBE Snapshot portal, and a simple majority by snapshot voting will be binding; the Multisig Guardian account will be mere executors of the snapshot votes’ decisions.

For flexibility and nimbleness in the initial development period, we propose that one of the FEI Team members (Joey, Seb, or Brianna) and defi3 would hold the multisig credentials. At some future point when the protocol is more mature we will transfer the multisig credentials to representatives duly elected by the TRIBE DAO. Through this process TRIBE DAO will exercise final control of this project in areas such as further endowing it with funds, transferring profits and changing basic parameters.

Funding:

To facilitate the successful launch and stability of FEI Lend we hereby request the following funds from either the DAO treasury, or the FEI Grants fund (subject to further discussion):

10,000,000 (10 million) TRIBE, and 15,000,000 (15 Million) FEI tokens as the starting capital of FEI lend, this will be deposited into the protocol upon launch to provide liquidity and stabilize initial interest rates. These funds will still be subject to the control of the TRIBE DAO by means of Snapshot voting. We have omitted ETH from the initial endowment because this is meant to encourage the use of FEI and TRIBE, we do not wish for the ETH lending facility to be highly liquid from the start.
250,000 time-locked TRIBE as an incentive for the working group, upon the completion of the protocol. We wholeheartedly believe in the future value of FEI protocol, therefore we wish to show our faith by only requesting time-locked TRIBE tokens as compensation. Having started the project on May 4, we anticipate a total expenditure of roughly 2200 man-hours. As suggested in earlier forum posts, we will abide by a base rate of $150/hr, though we feel that $300/hr is appropriate for the core solidity development. These time estimates account for the substantial redundancy in the programming process as outlined below:
30,000 TRIBE or equivalent funding towards our external auditor. As Joey has communicated to us the stressed state of their auditing resources due to the great surge of Defi projects, we have settled upon our internal auditing protocol. We will employ the “Pair Programming ‘’ method for all core protocol development. Our programmers will be organized into squads of three, with two principle pair programmers working with each other in real time; while the group auditor will step in at disputes for a final ruling.
Our current team composition would be:
Group [1] Frank Luo (PP-1), defi3 (PP-2), khbw (Audit);
Group [2] blueplanet (PP-1), taurus (PP-2), defi3 (A);
Group [3] potential developer from COMP team (PP-1), defi3 (PP-2), khbw (A).
KHBW will engage their in-house senior auditor, who will stay out of the project entirely and only step in at completion to start his independent audit. At the completion of the audit, he will present his credentials and audit report to the FEI Team for review. The FEI Lend team are significant holders of TRIBE in our own right, and we endeavour to achieve only the highest standard of integrity and safety for protocol funds; our primary motivation of creating FEI Lend is to bootstrap the use cases of FEI.

If all goes according to plan, we hope to deliver FEI Lend in an operational state consisting at least three pairs by the end of July 2021

Although sharing considerable similarities with other lending protocols, we have decided that it is absolutely necessary to build FEI’s own lending platform instead of merely integrating with others. Our consideration has been very thorough and we have concluded that this project would deliver value to the FEI community in the following respects:

As stated in the Preamble, FEI Lend seeks to immediately address the already existent demand of borrowing and lending FEI/TRIBE. Many community members either need to adjust and balance their staking positions, or need to cover other outside obligations without wanting to unwind their position altogether. Outside protocols such as RARI offer very limited liquidity and are not safe due to their extensive entanglements. FEI Lends will address this use case directly
Although it is likely that FEI might be accepted by many integrations, including by other lending platforms, in the immediate future. It is quite unlikely that TRIBE would be widely integrated elsewhere for a very long time. For instance COMP has integrated DAI stablecoin as an asset for many months now, but MKR is not integrated. This is also the case with VISOR v2, which has already unilaterally included FEI-ETH, but again unlikely to ever see TRIBE. FEI Lend will allow TRIBE holders to leverage their holding for uses elsewhere, while stimulating FEI demand at the same time.
FEI Lend will eventually allow FEI Protocol to control and lower the prevailing borrowing rate of FEI across the entire DEFI market. Per our vision of the future operation of FEI, (the creation of excess reserves is elaborated in greater detail at FEI's future as a decentralized central bank 5) FEI might one day be capable of automatically issuing excess liquidity above what is deemed to be the safe collateral ratio. Freshly minted, protocol controlled FEI can be deployed on demand through FEI Lend; these funds then may be used to lower the prevailing interest rate of FEI across the entire Defi ecosystem, and can be further loaned out to other platforms for various uses. Even without these future refinements to FEI, at the present moment, the lump sum endowments of FEI alone will already push the prevailing rate of FEI beneath that of other stablecoins. A consistently lower borrowing rate for FEI will make it more attractive than other stablecoin competitors** for diverse defi applications ranging from liquidity staking to flash loans; therefore increasing the organic demand for FEI. In the far future, these operations will allow FEI Lend to become the nexus of a sprawling network being able to influence the wider interest rate of Defi through the “FEI InterBank Overnight Rate” [FIBOR] system.
Creating a bespoke platform will vastly reduce long tail risks associated with other platforms. Allocating PCV funds to other lending platforms for interest rate subsidies would drastically increase the risk profile of PCV funds, for instance RARI has built a complicated Tranches system and pools all of their stablecoins together, while Sushiswap Kashi has over 95 pairs. On May 19, the largest lending platform of BSC network VENUS, was subject to a massive liquidation event as a group of attackers performed a pump and dump of its governance token XVS and borrowed large sums of ETH and wBTC at the very top of the pump. The subsequent liquidation has resulted in over $100M of losses for the platform. FEI Lend will seek to minimize this risk exposure, by severely limiting the available pairs, and lean on the extremely deep liquidity of TRIBE-FEI as a measure of security.

In the future FEI Lend can operate at very low or even zero spread between lending and borrowing. As FEI use cases expand, the lending platform fees saved by FEI Lend would become very considerable. This will allow FEI to control and subsidize its own prevailing interest rate while having total control of its risk profile. FEI Lend can even operate as a temporary loss leader to promote the usage of FEI. (idea courtesy of @joey) Ultimately, a great deal of the funds at FEI Lend may be loaned out to other institutions for redistribution after all, but our light weight intermediary allows FEI to exercise a higher level of control of its own interest rates; much like how the FED operates a desk in New York for its Open Market Operations.

White paper: FEI Lend