IndexCoop Fuse Pool on Rari Capital

[oneski22]

Owners: @oneski22/ @Mringz / @BigSky7 [Discord: TrotNixonLine, Mringz | Index, BigSky]

Channel: Business Development

TL;DR: Create a Fuse Pool on Rari Capital, controlled by IndexCoop to create a money market for our own products.

Proposal:

We are proposing creating an IndexCoop controlled Rari Fuse Pool.

Initially, the Fuse Pool will contain the following assets:

• INDEX
• DPI
• MVI
• ETH2x-FLI
• BTC2x-FLI
• USDC
• DAI
• ETH

The exact market parameters will be decided in conjunction with experts from Rari Capital and posted prior to a snapshot vote. Management of the fuse pool will be initially handled by Samuel Shadrach who leads fuse pool management for the Rari Capital team. The process will gradually be handed off to members of Index Coop.

Why Rari Fuse Pool:

Rari Fuse Pools will allow us to create money markets for our products and bring several key wins and abilities to both IndexCoop members and the IndexCoop DAO. A Fuse Pool is basically our own Compound-style market where the DAO will control which assets are listed, the interest models, collateral factors, etc.

Benefit 1: Operational and Investment Flexibility

The tokens in the pool are all assets held by our treasury. By enabling our operations account to borrow stables against our INDEX and DPI holdings we can avoid us having to sell at low INDEX prices, giving our treasury managers flexibility in how to manage our funds.

Additionally borrowing USDC against some of our INDEX holdings and placing them into a yield source such as Yearn (or even Rari’s Aggregator) would allow us to earn a return on our otherwise stagnant INDEX supply.

Lastly, our treasury earns DPI, MVI, and FLI from streaming fees. Instead of having these assets sit stagnant in the treasury, we will be able to supply these to the Fuse Pool to earn a yield until they are needed.

Benefit 2: Reduced Sell Pressure from Contributors

By providing a lending market for INDEX, contributors can now unlock the USD value of their earned INDEX without needing to sell their equity, reducing sell pressure and keeping our core contributors even further aligned with the project for the long haul.

Benefit 3: Added Extrinsic Productivity

We are providing our own venue for our products to gain extrinsic productivity. For some (MVI, INDEX, FLI) this will be the first time such an option has been available. High demand here could press other protocols (CREAM, AAVE, Compound) to list us.

Benefit 4: Alignment with a great partner DAO

The Rari team is excellent and well connected to many other DeFi projects. They are a DAO we have wanted to integrate with for a while. We have been holding off on doing a fuse pool integration until we found an angle that made sense to the Coop, and we feel this adds value to all stakeholders. This can be the first of several integrations between our DAOs. For example, we can use their yield aggregators as sources for IP down the line (or SYI should the methodologists deem appropriate).

Some KPIs to track success:

• Usage of the pools
• Decrease in the number of contributors dumping rewards immediately
• Return on stagnant DPI/MVI held in the pools

Risks:

Smart Contract Risk:

Rari Capital Smart contracts have been audited by Quantstamp and are presently under going a second audit.

Liquidation Risk:

We will be working closely with the Rari Capital team to ensure that the pool parameters are responsibility set to reduce risk of liquidations.

Added Context: Previous Post

Please comment below thoughts, questions, & things we may have missed.

[ansteadm]

Yaaaas! Been waiting for this. Can’t wait for this to go live. Need to lend out my INDEX

[jaibhavnani]

I’m so excited for this one, specifically the Fuse pools being able to underpin the Index Coop’s future products, like a flexible leverage DPI-2x

[Mringz]

I am a huge supporter of setting this pool up. Currently there are no protocols for me to borrow against my INDEX. That means that every time I need stables I have to sell. Which results in downward INDEX price pressure. I believe a lot of Index coop members have a sizeable INDEX position giving them the ability to unlock some liquidity without selling will be a huge value add. Secondly because we dictate the parameters of the pool it enables us to regulate the borrowing demand to reduce risks of liquidation. Super keen to see this set up.

[anon10525910]

@oneski22

Management of the fuse pool will be initially handled by Samuel Shadrach …The process will gradually be handed off to members of Index Coop.

curious, what exactly does it take to manage the pool?

[Erebels]

I am in 100% support of this. It would be awesome if we can borrow against our INDEX.

[oneski22]

From what I understand managing the pool means adding new assets and/or adjusting market parameters (CF, Reserve Rates, etc) as needed.

The processes and execution will be handed over to the Coop once we feel confident in our ability to manage the pool.

[mel.eth]

I’m not familiar with the plumbing of Rari pools, but if there’s a risk that one token failing could drain the liquidity of others in the pool I’d be a little hesitant (e.g. AMM style 1 goes to 0 they all go to 0); otherwise no issue there.

Management of the fuse pool will be initially handled by Samuel Shadrach who leads fuse pool management for the Rari Capital team.

A risk I see other than protocol is that you mention a third party will be managing the funds - I think that should be added to the ‘risks’ section of the proposal and discussed a bit (multisig?) along with a brief outline of how management of the pool will be handled post-handoff (maybe along with an anticipated timeline for the handoff). What I’m getting at, is that we’re going to have treasury funds that are on loan AND user funds tied up in this (and leverage - inherent risk), I think it’s important to understand whether we’re trusting people or code at each step of the process. If Rari have this process outlined already a link would be awesome.

By enabling our operations account to borrow stables against our INDEX and DPI holdings we can avoid us having to sell at low INDEX prices, giving our treasury managers flexibility in how to manage our funds.

I worry a bit about the the devil being in the parameters as it relates to IC borrowing against treasury - mainly how much leverage are we talking when borrowing stables against treasury assets and what yield sources are on the table? This is listed as a benefit - but it’s unclear if this proposal gives tacit approval to do those things or just highlights them as a potential benefit. Personally I’d like to see this proposal solely address establishing a pool and a separate discussion of how to make treasury assets productive (if that’s already happened and I missed it please link to it - I’m trying to get eyes on everything in the forums; it’s a journey).

Awesome work here - this is a major value unlock!

[oneski22]

Hi @mel.eth responding to these points:

if there’s a risk that one token failing could drain the liquidity of others in the pool I’d be a little hesitant

Rari Pools are essentially private forks of Compound. There is no IL risk associated with AMMs, the risk is that bad debt is issued and those who supply tokens can’t be made whole should they withdraw their tokens

a third party will be managing the funds

The IndexCoop will be in full control of the market at all times. We will simply be working very closely / relying on Rari’s expertise to set the initial parameters. All changes will need to be signed from IndexCoop’s designated address.

IC borrowing against the treasury

We are merely highlighting a potential benefit. This proposal does not authorize the treasury to borrow again its holdings or execute any leveraged strategies. I would expect the TWG to clearly articulate and receive DAO approval before executing any such strategy. This proposal is only for the creation of the fuse pool.

@Matthew_Graham, @prairiefi, and @AcceleratedCapital can speak more to how the Treasury would use such a market however I would read more about their visions for the Investment and Operations accounts below:

Goal and Philosophy of the Operations Account

Contributors: @Matthew_Graham @AcceleratedCapital @verto0912 @DarkForestCapital @MrMadila @dylan @ElliottWatts Background Continuing on the initial Treasury Management Goals and the subsequent TWG Treasury Diversification Proposal we seek to outline the Goal and Philosophy of the Operations Account. We expect this post will be the first of at least three forum posts that develop the creation of the Operations Account: 1. Goal and Philosophy (Why), 2. Composition (What), 3. Execution (How). A …

TWG - Investment Account Philosophy

Contributors: @Matthew_Graham @prairiefi @verto0912 @DarkForestCapital @MrMadila @dylan @george @ggraham1 @ElliottWatts Summary Now that we have established the Investment Account, it is important as a DAO to consider how we plan to manage the investment portfolio. To be most effective, we propose documenting objectives, constraints, and governance mechanisms to clarify responsibilities. Institutional investors typically do this by utilizing a standardized document referred to as the Inves…

.

[mel.eth]

Thanks for the lightning-fast reply! And thanks for the links; I’m a fan of any post with ‘philosophy’ in the title.
These responses cover my concerns. I’m in support. Y’all are legends. Have an awesome weekend!

[anon10525910]

gotcha, drilling in further, Is the process and execution a technical or non-technical one?

[Matthew_Graham]

Seeing as Rari is forking Compound, there are a few technical considerations I think we should be considered.

What Oracle feeds are being used ?

• Aave uses Chainlink
• Compound uses CoinbasePro & OKEx (Target getting DPI listing on OKEx to open up Compound … )

I am assuming Rari is using Uniswap price feeds. Some protocols have had issues using Uniswap price feeds when the size of pool using the price feed is greater than the pool it is tracking (Integral Resistance). INDEX is most liquid on Sushiswap, as is BTC2x-FLI. I don’t read about protocols using Sushiswap price feeds…

For lending platforms to be successful there needs to be sufficient on chain liquidity to enable the liquidation functionality to occur without distorting prices. There is good liquidity with DPI and chainlink oracles to support this which makes this product low risk. But MVI, BTC2x-FLI & INDEX don’t have a lot of on chain liquidity.

With limited on chain liquidity for some tokens - They really are not suitable as collateral. Linking in the Compound documentation here and providing some extracts below.

If BTC2x-FLI, MVI & INDEX are listed as collateral, then I have a lot of questions around how Rari is assessing these risks. Having just gone through this process for DPI with Aave, I would not encourage using a low liquid, high vol assets as collateral - to me that is asking for trouble and the loan would need to be massively over collateralised. For example, if someone dumps on a low vol asset, say INDEX, the price movement will be substantial. If Rari is using an oracle feed from the pool that was just dumped in, it could trigger a liquidation event for everyone using INDEX as collateral.

I understand the proposal is saying Rari will be managing this aspect for Index Coop - I think we need to do some due-diligence here and highlight why something can be listed on Rari but not Aave or Compound.

I must apologies here as well. I did have chats earlier in the week about this proposal and it is not until I went back through the risk assessment documentation on Aave and reviewed the Compound documentation that the above thoughts came to me.

Fantastic effort and great idea. I think we are early on some assets here, but DPI and INDEX (post Balancer Pool) have potential. If we are serious about doing this, I would like to go very deep on the associated risks. To me this would be a minimum that Index Coop needs to do in order to promote or endorse the concept.

[mel.eth]

@Matthew_Graham - Would the risks you mention not be substantially mitigated by reducing the collateral factor to very low levels (or fully mitigated by setting to zero) to start for the low liquidity tokens? This would still allow deposited assets to be borrowed (unlocking some leveraging strategies) while managing risk on behalf of depositors. As I see it, the concerns you raise are super valid and need to be explored further to determine collateral factors, but within the proposed framework fall under the umbrella of manageable risk as opposed to untenable risk given my current understanding.

[oneski22]

Hi Greg, non-technical, there is a UI that a whitelisted address will be able to access. If we want to heavily customize there will be collaborative technical work with Rari Team, however as of now nothing we would want to do short to medium term is outside of the current support functionality.

[oneski22]

Updating on Pool Composition:

For launch the pool would be exclusively:

DPI
USDC
DAI
ETH
INDEX** (pending check that we have the right oracles)

Eventually we would add:
MVI & FLI suite, with collateral usage disabled. (again pending oracles)

Chainlink is the preferred oracle (and now the oracle solution of Compound as well), however they do support UniV2 and V3 TWAPs.

As for the “Rari will be managing”, Rari will provide recommendations, however as the owners and controllers of the pool we are the ultimate decision makers of what the various parameters are and can change as necessary.

Lastly as for why something can be listed on Rari but not Aaave or Compound, Rari is a permissionless pool. We own the Fuse pool, decide the assets etc. Its a private Compound style market.

[jdcook]

@oneski22 @Mringz @BigSky7

So the proposal is to seed the pool with portions from our treasury? Am I understanding that correctly? Do we have numbers around how much we would need to seed with to make this effective?

Aside from the time and effort to manage the Fuse pool, are there any other costs to the Coop - especially $ costs?

I am trying to get up to speed on the ins and outs of Fuse pools - it seems like our products (DPI, INDEX) would carry very little borrowing demand… is that an issue? I know that was discussed a bit in the original proposal.

I want to see this happen!

[oneski22]

Hi @jdcook this proposal does not involve the movement of treasury funds.

We will be starting a new thread with the final proposal shortly. It is simply to spin up the pool and bring it under the control of IndexCoop.

[jdcook]

Ok, I will wait to see the final proposal to follow up on my questions! Thanks.

[Matthew_Graham]

After reading through the chats here, I am wondering will Rari Fuse Pool move the dial. Is it worth the effort of doing the risk analysis, determining collateral ratio and designing a model that incentivises liquidations to occur.

I’m particular interested to see if liquidations get called. We all seen what happened with the FLI products via Compound (Tier 1) when the market was volatile. We have the same risk exposure on Rari (Tier 3 IMO), but the pool is a smaller and doesn’t have the presence of Compound in the market.

If the liquidation are not performed efficiently, then the loans risk being under collateralised. Unlike Aave; Compound & Rari don’t have a safety module to make lenders whole with in the event the liquidator function is not performed and there is a loss of capital. Given Rari doesn’t have a large safety net behind them at the DAO level and has recently been exploited, I’m not sure if the DAO could make lenders whole if something went wrong. It would probably require outside funding.

This liquidation risk is a blocker for me that would prevent Index Coop capital from being deployed into Rari. It would be high risk and it is unclear what the Return On Investment (ROI) would be.

I’d be really curious to see what a peer comparison would bring. I understand the narrative, but what does the data say.

• Why Rari and not another protocol
• Why Rari with Cream and Aave
• What data is there available from RAI that can be used to determine if this will have any impact on our KPIs and NorthStars
• What is the ROI

Aave (Aave backstop, high barrier to entry, no upkeep by Index Coop)

Rari (Compound, fragmented markets limiting cross asset borrowing options, Index Coop involvement required)

Cream (Compound with no barrier to entry, Cream manages everything)

[overanalyser]

1. as with @Matthew_Graham I am wary of the costs involved in managing the pool, however, I am intrigued by the idea.
2. I suspect I’m thinking of using the pool in a different way…

The proposal implies that the coop should be depositing INDEX / DPI and borrowing stable coins etc.

I don’t like that idea, as it forces our treasury to become a more active trader and we would need to watch the risk of liquidation and spikes in interest rates when stable coin utilisation is high. I do not consider such active trading a desirable core competency for the coop treasury.

I would approach this pool in a different way:
a) the coop deposits stable coins and allows others to borrow them
b) we set the ETH parameters (collateralization ratio) to discourage ETH deposits and so encourage the use of DPI / INDEX as collateral.
c) we don’t allow INDEX borrowing - as this can be used in governance attacks.

This would effectively mean that the coop would be lending USDC to DPI / INDEX whales in return for interest paid in USDC.

By being a stable coin lender (and not borrowing anything) the coop risks would be reduced significantly [There is always the risk that a price crash would result in a borrower being liquidated and the liquidation fails to repay the borrowing - as happened to Maker in March 2020]

Downsides for the coop would be that:

1. USDC interest received might be less than the coop could get on yearn / AAVE etc (if we plan to do that)
2. We are taking on smart contract risks/liquidation not recovering the full value borrowed
3. INDEX locked up and unable to vote.
4. we will need to be involved in the Rari value parameter setting

Benefits:

1. We make INDEX useful to people who want to be active borrowers.
2. We open up another place to deposit DPI to borrow stable coins (alongside Cream)