SDIP-28: NFT Lending Expanded Program

NiftyFifty

Brief Authors: glory, NiftyFifty
Project lead: glory, NiftyFifty
Squad: Treasury, Acquisitions
Creative approval: N/A

Abstract

After a successful execution of our NFT Lending MVP (see “Motivation” section for details), we propose that Treasury and Acquisitions keep working together doing Squiggle-backed loans on NFTfi with an expanded budget of 200 ETH (if Gnosis multisig wallets are still not in production on NFTfi, we propose a reduced budget of 100 ETH).

By offering loans collateralized by Squiggles, we accomplish the following primary goals:

Acquire more squiggles (if the loans default) - integral part of SquiggleDAO’s mission
Generate some yield on our -otherwise idle- ETH (if the loans get repaid)
Build our brand (by positioning SquiggleDAO as the main lender for Squiggles)
Help the Squiggle (the more liquid the NFT, the more valuable)
Help the community (by having liquidity available at reasonable rates)

This is another step towards our long term goal of becoming the primary liquidity provider for Squiggles.

Motivation

Last April we voted and approved SDIP-22: NFTfi Lending / Borrowing Program. One of the primary goals was for this to serve as an NFT Lending MVP in anticipation of launching an expanded lending program.

As Gnosis vaults were still not in production on NFTfi, we used a single-sig hardware wallet (controlled by the Treasury Squad) with a lending budget reduced to 25 ETH.

With the 25 ETH, we did 4 loans taking into account all the Squiggle nuances (not only traits, but also day zero, double-digit, color spread, aesthetics, etc.). In addition, we adjusted the LTVs based on the moment of the cycle (i.e. it is not the same doing an 80% LTV loan when floor is around 5 ETH vs 15+ ETH), we are more aggressive when we are near the bottom of the range and would be significantly more conservative after a quick run up on Squiggle floor prices.

All loans were done with a maturity of 30 days and at 20% APR. Now all the loans have been repaid and we generated a 2% ROI and a 22% IRR on average.

One of the limitations we identified was that making loan offers on NFTfi is a time intensive process, hence our Interim Developer (Arkaydeus) is working on a bot to automate it.

We are proposing an expanded NFT Lending program as we have more 850 ETH on our Treasury and Squiggle backed loans fit us perfectly. If the loan defaults, we acquire a Squiggle at a discount (one of our primary goals as a DAO). If the loan gets repaid, we generate double digit yield (NFT lending provides the best returns in ETH at the moment). Also, it is important to mention that we are not using any ETH for acquisitions to be conservative with our Treasury assets in a bear market.

Finally, NFTfi has already confirmed an airdrop and this program will allow us to increase the probability of being included in it.

Success Conditions

Issue 10 loans successfully
Participate in the NFTfi airdrop
Earn at least 10,000 USDC equivalent in NFTfi tokens during the airdrop
Experiment with different loan terms (maturity, APRs, LTVs, etc) to learn the preferred borrowers’ choices
Track data required to expand the program (unfulfilled loan supply, trackrecord, ETH utilization, max ETH loaned, etc.)

Specification

One 2 out of 3 Gnosis wallet will be set up for this activity (NFTfi Lending wallet) - proposed signers: Gregg, glory, and NiftyFifty
- Lending wallet will be funded with ETH from Treasury
  - If Gnosis wallets are still not in production, we propose to use a single-sig hardware wallet (controlled by the Treasury Squad) with the consequent reduced budget (as per “Abstract”)
If an NFTfi airdrop does occur, a vote will be held in the “Ongoing Votes” channel in Discord to determine what to do with the tokens (sell, keep, etc.)

Launch Details

Project will be launched according to the specifications above. We will launch a marketing campaign (ideally with NFTfi) to make sure the community is aware of our program.

Costs

200 ETH (though some or all of this could return back to the treasury depending on the outcomes of the lending activity).

Rewards

N/A

0xgregory

Correct me if I'm wrong, but we have confirmation of no Multi-sig support?

So 100 ETH allocated. And does that mean we'd still be setting up a 2/3 for this execution? If not, how would we operate this?

Are we tracking loan data somewhere? Perhaps we should link this here as well.

If there are more loans available than capital available, how will we choose which loans to service? What's the criteria that we have developed, or need to develop, in order to max performance?

Are there any risks that we need to identify? eg. smart contract risk, volatility, etc.

NiftyFifty

Thanks for your comments, 0xgregory

Currently NFTfi doesn't support multi-sig, but it's in their roadmap. They deployed their v2 set of smart contracts in early April. It is a modular system which allows them to slowly deploy new features and multi-sig is in their plans.

Initially it would be 100 ETH and if they release the gnosis integration, we would expand (assuming this proposal gets approved) to 200 ETH. Until gnosis integration, my suggestion is to keep operating as we've done so far (from a single-sig hardware wallet controlled by glory)

We are tracking loan activity (with some proprietary data) so prefer not to share in the forum - happy to discuss on Discord and/or TownHall. But we can share our NFTfi account.

In the case of excess supply, we would adjust our offers to be a bit less competitive (in terms of LTVs and APRs). In fact, after the current run up, we are adjusting LTVs to around 60% (from 80% when floor was at 5 ETH).

These are the risks involved with this strategy:

Smart contract risk: the most important risk we are assuming. It is limited as there has been more than $200M of loan volume and more than 10,000 loans with no issues on NFTfi (that includes v1 and v2). NFTfi v2 has been audited by Chainsecurity and Halborn and it's been in production since early April without issues.
Market risk: if the floor drops below our loan value during the life of the loan, we would be acquiring a Squiggle at a price above market.
Concentration risk: we will monitor who are the counterparties and will try to limit the exposure to any particular counterparty (or modify our offers to reflect this risk).
Credit risk: obviously limited by the collateral.

It is important to mention that, in addition to the above risks, there is an opportunity cost as we will be using some ETH, but this is limited as -at this stage- we would not be using that money otherwise.

cheesebiz

I support this proposal. The first trial of lending was a success, and this will allow us to generate some revenue while also supporting the squiggle market. Some thoughts on the proposal below:

1) multi-sig would be ideal. 200 or even 100 eth is a lot for one person to be in control of. Not that I don’t trust the team but it is a risk for the DAO. If there is no other way around we just need to make sure we are okay accepting this risk.

2) In case of excess supply, I agree w/ Nifty that we can likely obtain more favorable loan terms. My bias would also be to lend to the more rare and higher valued squiggles because in the event of foreclosure we’d get an asset that is more difficult to purchase.

3) What about if our capital is exceeds the amount of squiggles seeking a loan? Would we want to consider lending to other blue chip projects (like maybe only punks), at more conservative terms? Just a thought if we have a significant amount of idle eth.

4) re the bot - will the bot be able to incorporate the more nuanced traits? Or will there still be a need for manually making certain loans? What is the process for testing the bot before it goes live, and what is the process for monitoring the bot for errors / wrong bids?

NiftyFifty

cheesebiz Thanks for your comments, cheese!

1) we are in conversations with NFTfi so they know that multi-sig is important for us and will allow us to expand our budget. Hopefully they'll implement it soon

2) I like the idea of going for the rarer in case of excess supply

3) haven't considered it. So far it was possible to use all our funds and looks like -even with the expanded proposal- we should still be able to. But something interesting to consider for the future

4.1) we can incorporate traits to the bot, but we'll keep it simple to start (to minimize risk). Maybe just normal-slinky-fuzzy-etc. to start. Also our plan is to manually put loan offers as well (so we can fine tune for things like aesthetics, day-zero, etc. and experiment with different durations)
4.2) we can test it in Rinkeby and our idea is to slowly deploy funds, i.e. start with 5-10 ETH and slowly build up from there

cheesebiz

NiftyFifty awesome, thank you for the response and all the info. All sounds great.

0xgregory

What a great, thorough conversation. Thanks for putting in the effort!

Long term, this is a no brainer for the squiggle and for SquiggleDAO. Glad to see that we're taking a measured approach to becoming experts here