YIP 77: Bootstrapping Yam Fuse Pools

Yam Dao has been whitelisted as a ‘Fuse Pool Creator’:
https://snapshot.org/#/rari/proposal/QmdNYSDUdojyAREMf2hzf2nEw7iRTSbvrSsVr4hWX2LHG5

Rari Capital’s Fuse product, allows for the creation of custom lending/borrowing markets which are parameterised by the pool creators, in our case this would be Yam Dao. By creating a Fuse pool centered around the YAM token the community can reap the following benefits:

Increase token utility:

Borrow stablecoins against YAM collateral
Borrow YAM or YAM/ETH SLP against stables to farm without price risk
Borrow YAM or YAM/ETH SLP against stables to hedge YAM positions

Make YAM a yield bearing asset:

Existing YAM or newly minted YAM would be yield bearing if deposited into a Fuse pool maintained by Yam.

Earn yield on a portion of treasury assets:

Any Treasury assets deposited into the pool would also become yield bearing

Earn Fuse Pool admin fees:

Rari have agreed to pay 20% of the initial Fuse Pool admin fee to the Yam treasury (this fee is charged at 10% of interest earned within the pool)

Benefits aside, Fuse pools come with standard lending/borrow market tail risks that are common across Defi, some examples:

Oracle manipulation (TWAP’s are used, increasing the risk and cost of performing this attack)
Imperfect liquidation system in times of extreme volatility

Yam Dao has engaged with Rari to advise on setting up these pools to be as secure as possible. It is noteworthy that to date no Fuse Pool has been exploited.

Pool Specifications:

The initial Fuse Pools would consist of the following assets:

YAM
USDC
DAI
YAM/ETH SLP

As YAM and YAM/ETH LP are lower liquidity tokens and more sensitive to large price movements, two separate pools will be created to isolate this risk. The first and larger pool will include all assets listed but only allow the lower liquidity tokens as collateral, and a smaller second pool consisting only of these two tokens where they are both borrowable and allowed as collateral will be created. Both pools are necessary to reap all the benefits listed above.

Action Required:

To create a liquid lending/borrow market for Yam token holders in the form of a Fuse Pool, bootstrapping liquidity is necessary. If we consider Yam Dao’s asset base to be treasury funds + contributor vesting pool’s YAM, we arrive at the figure of $5.36M + $907K = $6.27M at the time of writing. It is initially proposed that we deposit 8% of Yam’s asset base into the initial Fuse pools.

This is enough to create a liquid market, but not too large a sum that in a black swan event that we haven’t seen before would severely impact the treasury. 8% of the current asset base would amount to $500k. The full amount would become interest earning once borrowers take out loans, and Yam dao can chose to withdraw these funds at any time in the future (less current amounts loaned out which can be withdrawn once repaid)

It is important to note that a large portion of this amount would be YAM. Which can be sourced from the existing YAM in the contributor vesting pool or can be sourced by minting new YAM. Please note that this choice is included in the voting options.

The non YAM portion would need to be sourced from the treasury. If this proposal is to be implemented, these funds would be secured by using existing treasury assets and selling Sushi and DPI tokens, as they are the most outsized positions relative to the rest of the treasury as indicated in the July treasury report.

Polling
There is a live snapshot vote here: Snapshot
It runs until 3:59 PM UTC on July 22nd.

This Vote has Passed :partying_face:

Adding information on the Fuse pools here (Info from @THEVDM1)

Fuse Pool Params:

Pool 1: Yam: Jewel

Designed for casual users looking to borrow stables/eth against yam without being exposed to more pool risk and higher liquidation incentives than necessary. The key risk reduction is that as Yam is not borrowable the pool is not exposed to an insolvency event triggered by a violent increase in the Yam price within the TWAP window. An example of an event that could trigger this is something like a coinbase listing.

Pool 2: Yam: Vardaman

For a more advanced user who can carry out the standard lending above as well as a wider range of strategies like beta neutral farming/short selling at a higher risk (the low liquidity assets become borrowable).

As Yam is borrowable here, the pool is more vulnerable to violent price increases.

The risk division also helps guide the treasury in capital bootstrapping/ investing in the pools, allowing us to allocate funds with a choice of two risk tiers.

Pool wide params:

Pool Close Factor Reserve Factor Liquidation Incentive
1. Yam: Jewel 50% (compound default) 0% * 10%
2.Yam: Vardaman 50% (compound default) 10% 20%**
  • no reserve factor applied as pool is less risky already by disabling borrowing low liquidity tokens

** liquidation in yam borrowable pool is high to account for high slippage incurred from Yam’s low liquidity

Specific params:

CF’s here were guided by ghosts_in_the_code from Rari who works on their risk management side.

1.Yam Jewel Collateral Factor Int. rate model
Dai 65% Jump rate stable/major
Eth 65% Jump rate stable/major
Yam*** Collateral only 50% Jump rate gov
2.Yam Vardaman Collateral Factor Int. rate model
Dai 60% Jump rate stable/major
Eth 60% Jump rate stable/major
Yam 50% Jump rate gov
Yam/Eth LP 50% Jump rate gov

No additional fees to be charged other than the agreed split admin fee applied on interest earned (of the 10% fee 8% to Rari 2% to Yam dao)

Pro’s and Con’s of 2 pools vs a single pool

Two Fuse Pools:

PROS CONS
Risk segmentation for users and treasury Liquidity spread across two pools
Separation of liquidation penalty (10% vs 20%) Confusing for users (could educated his away - docs, articles etc)
Less vulnerable to declining Yam liquidity

One Fuse Pool:

PROS CONS
Liquidity all in one pool No risk segmentation
Easier to understand for users (less education needed) No separation of liquidation penalty (10% vs 20%)
Vulnerable to declining Yam liquidity

Bootstrap Capital Allocation across pools:

Assets to be sourced from existing assets + sale of Sushi/DPI as per Krugman recommendation

YAM: Jewel % of $500k $ Amount
YAM 1% 5,000
DAI 50% 250,000
ETH 10% 50,000
Pool total 61% 305,000

As Jewel is a pool for users to deposit YAM and borrow DAI or ETH, it is the user that is depositing the Yam so this pool doesn’t need YAM liquidity as part of the bootstrapping effort. Also more capital is allocated to DAI than ETH in the above pool as the assumption is that there will be more demand to borrow DAI against YAM than ETH against YAM. Fuse Interest rates have guided this assumption.

YAM: Vardaman % of $500k $ Amount
YAM 27% 135,000
YAM/ETH LP 10% 50,000
DAI 1% 5,000
ETH 1% 5,000
Pool total 39% 195,000

As Vardaman is intended for borrowing of YAM and YAM/ETH, the user is depositing DAI and ETH so this pool doesn’t need DAI and ETH liquidity as part of the bootstrapping effort.

Allocations for these pools was made based on the following assumptions:

Order of demand highest to lowest for the 3 main Fuse pool strategies:

  1. Simple borrowing against YAM
  2. Hedging YAM
  3. Farming YAM with YAM/ETH market risk free

On top of this there is roughly a 60/40 split across pools with the safer pool being allocated the lionshare of 60% and the risker pool 40%.

Suggested game plan:

Go ahead with what was proposed in the snapshot with two Fuse pools as it is the more conservative approach that caters for a potential decline in Yam’s liquidity and violent price swings to the upside (done). Once the pools are live they can be monitored and if we feel one pool is superior relative to the observed risks, we can move liquidity from the one pool to the other over time if need be.