[DRAFT] Treasury Diversification Authorization Program - v2

[sky]

[DRAFT] Treasury Diversification Authorization Program - v2

Following the ongoing success of the first Treasury Diversification Authorization Program according to the Treasury Diversification Authorization Proposal executed Jan 23, 2021, DXdao is on it’s way to a more diversified treasury.

So far, you can find a summary here: https://gateway.pinata.cloud/ipfs/QmfZdBNMtEPteG7JuTqxbxg4Zo4o4veWcenwoJs6PrsPZC

Currently $2m of USD stables (mainnet)
Proposals in process for $1m more. (mainnet)
Another $450k of USD stables on xDai.
Total of $3.45m of USD stables

There remains a need to continue to diversify DXdao’s treasury into USD stable tokens and Yield earning USD stable tokens. This would continue to help secure a runway for DXdao to achieve its goals.

The proposal outlines a plan for DXdao that is split up into three objectives:

1. Acquire another $5m of USD-stable assets, using the three methods proposed below
2. Of all the USD assets DXdao hold, put $4m of these USD-stable assets into yield-earning protocols (this can be from existing and newly acquired holdings)
3. Explore ETH yield and other yield opportunities with up to 1,000 ETH - This can be ETH2 staking or ETH lending or ETH vaults.

The $5m additional USD-stable assets would currently be an additional about ~10% of DXdao’s treasury, maintaining large exposure to ETH price risk.

DXyield: DXdao will be seeking yield with its USD stable tokens.

This proposal authorizes use of Compound, Aave and Yearn protocols.

The first yield earning protocol for DXdao would likely be Compound protocol followed by Aave and/or Yearn.

Any other community suggestions would be welcomed via separate investment memo and Proposal.

The breakdown of the stablecoins shall be slightly adjusted to target 40% DAI, 30% USDC, 20% sUSD, and 10% USDT. In dollar amounts that is:

2,000,000 DAI
1,500,000 USDC
1,000,000 sUSD
500,000 USDT

For reference, you can find some yield reference rates here: Crypto Lending Rates - Earn Crypto Interest by DeFi Lending

There is currently no defined plan to hold non-USD based stable assets, but this topic should be looked into and addressed via alternative forum post and proposal.

Methods:

Again the most trustless method for DXdao to rebalance its treasury is to use the Gnosis Protocol Relayer that DXdao will soon be connected to via the Multicall Scheme. This should eventually be the primary, and perhaps only, way in which DXdao rebalances its treasury. However, the multi-call proposal will take a couple weeks to pass and then each relayer proposal will take roughly one additional week and also it would make sense to ramp up the size of the relayer trades somewhat gradually. Therefore, it could take more than a couple months to move a significant portion of the treasury into stablecoins by using the Gnosis Protocol Relayer alone. With ETH nearing its all time highs, it is important for DXdao to accelerate the treasury diversification process. This proposal authorizes diversification through the following 3 means:

1. Gnosis Protocol Relayer

When the Multicall scheme is installed, the Gnosis Protocol Relayer will become available for use. While it will make sense for DXdao to start small with this newly deployed system, this method is the most trustless, and this authorization proposal places no limits on the amount traded via the Gnosis Protocol Relayer towards the listed conversion target of $5M.

2. DXdao Developer MultiSig

This proposal authorizes any amount towards the listed conversion target of $5M to be traded via trusted multisig, granted that no more than 500 ETH is transferred to a trusted intermediary in a 2 day period, that the multisig executes trades within one day of receiving each disbursement, and that the multisig currently requires signatures of at least 3 of 7 addresses, owned by distinct REP holders with greater than 1% REP that have verified ownership of their multisig member address by signing a message with the multisig member address and including this message in a Keybase saltpack signed message along with the address that holds their REP and presenting this verification in an alchemy proposal (or the same REP wallet). Note that some REP holding addresses are Gnosis Safes which unfortunately lack the ability to sign messages and also can’t be members of the multisig, and therefore, this verification process relies on keybase profile identities.

3. “Member Balancer” Trading

The “Member Balancer” approach incentivizes third parties to send stablecoins to the treasury and ask for ETH in return. This allows DXdao to diversify its treasury without having to trust an intermediary, but does require the third parties to trust DXdao and for DXdao to provide some incentive. This proposal authorizes that 1.01 times an amount of ETH may be requested via alchemy proposal by a third party granted that they have traded that amount of ETH for stablecoins on a highly liquid DEX and transferred the stablecoins to DXdao no earlier than 2 hours before the proposal. If the proposer is not trading ETH for said stablecoins, for example because they already are holding the stablecoins, they may use the fair market value of ETH as reported by Coingecko at the time of request granted that the transfer of stablecoins has been made no earlier than 2 hours before the proposal. Furthermore, this proposal limits the amount that may be traded via the “Member Balancer” approach to $2M and requires that a minimum size trade per proposal should be 30 ETH.

Thoughts and Feedback welcome.

[Arhat]

We could incentivize member balancer trades to go through the weth/xdai pool on Swapr with gas reimbursement in vested DXD and then some.

[Powers]

I think this is a great post and curious to hear others thoughts.

The three objectives that Sky laid out (acquire $5m additional stable assets, put $4m into yield-earning protocols and explore ETH yield) are all apart of the next stage of “climbing the risk ladder”.

A couple other points:

• We should execute as many as possible through the Gnosis Protocol Relayer option. This is the most trustless and we should try to use this option whenever possible.
• sUSD has yield opportunity and diversification but it’s not been very liquid. USDT is about 100x as big as sUSD. We should see if the multisig still has problems executing it. What’s more, we will have to think of liquidity on GP.
• How to ‘access yield’ - the new multicall scheme will include a 14 day vote period for any contract, but we’ll have to figure out how to get into Compound/Aave/Yearn. We could purcahse cDai/aUSDC but I imagine we will have to come up with a system
• I really like the idea of the “Investment memo” for other investments that don’t fall into these categories. This would allow us to streamline these common (and relatively safe) diversification efforts but also open up to additional areas but require more diligence. I guess we could probably do this for all DXventures investments too?
• I’d probably be in favor of $6-7m+ in yield earning assets (as opposed to $4m). I don’t think we should be holding that much in stablecoins not earning a yield, especially as we’re handling worker payments on xDai
• Perhaps we should put a goal or date to accomplish by? 10 weeks after proposal passage?

[Arhat]

@Powers

How to ‘access yield’ - the new multicall scheme will include a 14 day vote period for any contract, but we’ll have to figure out how to get into Compound/Aave/Yearn. We could purcahse cDai/aUSDC but I imagine we will have to come up with a system

Yearn is contract-friendly. One can automate withdraw & send function with gnosis safe.

[JohnKelleher]

Few comments:

1. I think 500 ETH per two days is too high of a limit for the multisig. The previous limit was 100 ETH. I think 250 ETH would be reasonable as an incremental increase.

2. I think this proposal should be made on mainnet because of the amounts involved and to be consistent with the previous authorization proposal. In order to save on gas costs, the first transfer could be included on the same proposal. This proposal authorized the xDai base for signal proposals but it isn’t strictly clear whether all signal proposals should happen there. Ultimately mainnet has the authority and it would be good to set clear precedent for a future where activity on xDai could lead to a divergence in REP makeup compared to mainnet. That said, one of the attractive things about the xDai base is reduced voting costs. In order to encourage voters to express an opinion, we could do a mirrored proposal on xDai. Here is a suggested paragraph that could be included in the proposals:

“This proposal is being duplicated on xDxdao in order to enable people an affordable way to express their opinion. Mainnet takes precedent, but any discrepancy in the voting outcome would be informative for DXdao governance about the health of the voting/consensus process.”

3. Due to price volatility the ETH sent to the multisig in these two previous proposals: 13 and 14 will not be fully converted to stablecoins. In order to save on gas costs, would be good to include a note in this authorization proposal indicating that the ETH already in the multisig can be used for towards the diversification outlined in this proposal.

[Powers]

Just to visualize it a bit better. Here is what the treasury would look like (includes xDai, but excludes buyback reserve)

Of course doesn’t include things we’ve allocated for, like $1m for DX Ventures or additional funds for Swapr. And the DXD is only what is in the treasury (23k). There’s another 76k in the vesting contract.

[Iam_kmkm]

thanks for the easy-to-digest visualisation!

I have a naive question.

Looking at the treasury, I see among “Other” (total of 393k):

1. Pinakion
2. DMM: Governa…
3. OWL Token, and
4. Panvala pan

I have never heard about these tokens/project, and I don’t know their history in the context of DxDAO. So the question is:

Question: why do we prefer to sell ETH instead of these ‘altcoins’ first?

thanks!

[sky]

After incorporating community feedback, here is a revised draft to check out:

Signal Proposal: Treasury Diversification Authorization Program - v2 (Revised Draft)

Governance Note:

Given the sizable USD amounts involved in this Signal Proposal, this proposal is being submitted on mainnet DXdao.

This proposal is also being duplicated on xDXdao (xDai Chain) in order to enable REP holders a more affordable way to provide signal. Mainnet takes precedent, but any discrepancy in the voting outcome would be informative for DXdao governance about the health of the voting/consensus process.

Proposal:

Following the ongoing success of the first Treasury Diversification Authorization Program according to the Treasury Diversification Authorization Proposal executed Jan 23, 2021, DXdao is on it’s way to a more diversified treasury.

So far, you can find a summary here: https://gateway.pinata.cloud/ipfs/QmfZdBNMtEPteG7JuTqxbxg4Zo4o4veWcenwoJs6PrsPZC

Currently $2.85m of USD stables (mainnet)

Another $450k of USD stables on xDai.

Total of $3.3 of USD stables

There remains a need to continue to diversify DXdao’s treasury into USD stable tokens and Yield earning USD stable tokens. This would continue to help secure a runway for DXdao to achieve its goals.

The proposal outlines a plan for DXdao that is split up into three objectives:

1. Acquire another $5m of USD-stable assets, using the three methods proposed below
2. Of all the stablecoins in mainnet treasury, put $5m of these USD-stable assets into yield-earning protocols (this can be from existing and/or newly acquired holdings)
3. Explore ETH yield and other yield opportunities with up to 1,000 ETH - This can be ETH2 staking or ETH lending or ETH vaults or liquidity mining.

Details on 1): Acquire another $5m of USD-stable assets, using the three methods proposed below

The $5m additional USD-stable assets would currently be an additional ~10% of DXdao’s treasury, maintaining large exposure to ETH price risk.

For reference, assuming an ETH price of $2,500, the $5m would be 2,000 ETH. To make this authorization clear, the proposal would authorize the conversion of up to 2,000 ETH into USD-stable assets.

The targeted breakdown of the stablecoins shall be:

40% DAI → 800 ETH into DAI

40% USDC → 800 ETH into USDC

10% sUSD → 200 ETH into sUSD

10% USDT → 200 ETH into USDT

ETH price moves will affect the exact end percentages, but individual proposals that action the movement of funds will outline what exactly should be executed.

Important parallel note:
Due to price volatility the ETH sent to the multisig in these two previous proposals: #13 and #14 will not be fully converted to stablecoins. In order to save on gas costs, this authorization proposal allows for the ETH already in the multisig to be used towards the diversification outlined in this proposal.

Details on 2): Put $5m of these USD-stable assets into yield-earning protocols

DXdao introduces the idea of DXyield: Given that DXdao will be holding a sizable position in USD-stable assets, DXdao will be seeking yield with its USD stable tokens.

This proposal authorizes use of Compound, Aave and Yearn protocols.

The first yield earning protocol for DXdao would likely be Compound protocol followed by Aave and/or Yearn.

Any other community suggestions would be welcomed via separate investment memo and Proposal.

For reference, you can find some yield reference rates here: Crypto Lending Rates - Earn Crypto Interest by DeFi Lending

There is currently no defined plan to hold non-USD based stable assets, but this topic should be looked into and addressed via alternative forum post and proposal.

Details on 3): Explore ETH yield and other yield opportunities with up to 1,000 ETH

The goal here is to earn yield on ETH that DXdao holds. Any community suggestions would be welcomed via separate investment memo and Proposal.

Methods outlined:

Again the most trustless method for DXdao to rebalance its treasury is to use the Gnosis Protocol Relayer that DXdao will soon be connected to via the Multicall Scheme. This should eventually be the primary, and perhaps only, way in which DXdao rebalances its treasury. However, the multi-call proposal will take a couple weeks to pass and then each relayer proposal will take roughly one additional week and also it would make sense to ramp up the size of the relayer trades somewhat gradually. Therefore, it could take more than a couple months to move a significant portion of the treasury into stablecoins by using the Gnosis Protocol Relayer alone. With ETH nearing its all time highs, it is important for DXdao to accelerate the treasury diversification process. This proposal authorizes diversification through the following 3 means:

1. Gnosis Protocol Relayer

When the Multicall scheme is installed, the Gnosis Protocol Relayer will become available for use. While it will make sense for DXdao to start small with this newly deployed system, this method is the most trustless, and this authorization proposal places no limits on the amount traded via the Gnosis Protocol Relayer towards the listed conversion target of $5M.

Note: If Gnosis Protocol is no longer active, other methods that DXdao can execute itself should be explored in place of Gnosis Protocol.

2. DXdao Developer MultiSig

This proposal authorizes any amount towards the listed conversion target of $5M to be traded via the DXdao Dev multisig, granted that no more than 250 ETH is transferred to the DXdao Dev mutlisig in a 2 day period, that the multisig executes trades within one day of receiving each disbursement, and that the multisig currently requires signatures of at least 3 of 7 addresses, owned by distinct REP holders with greater than 1% REP that have verified ownership of their multisig member address by signing a message with the multisig member address and including this message in a Keybase saltpack signed message along with the address that holds their REP and presenting this verification in an alchemy proposal (or the same REP wallet). Note that some REP holding addresses are Gnosis Safes which unfortunately lack the ability to sign messages and also can’t be members of the multisig, and therefore, this verification process relies on keybase profile identities.

3. “Member Balancer” Trading

The “Member Balancer” approach incentivizes third parties to send stablecoins to the treasury and ask for ETH in return. This allows DXdao to diversify its treasury without having to trust an intermediary, but does require the third parties to trust DXdao and for DXdao to provide some incentive. This proposal authorizes that 1.01 times an amount of ETH may be requested via alchemy proposal by a third party granted that they have traded that amount of ETH for stablecoins on a highly liquid DEX and transferred the stablecoins to DXdao no earlier than 2 hours before the proposal. If the proposer is not trading ETH for said stablecoins, for example because they already are holding the stablecoins, they may use the fair market value of ETH as reported by Coingecko at the time of request granted that the transfer of stablecoins has been made no earlier than 2 hours before the proposal. Furthermore, this proposal limits the amount that may be traded via the “Member Balancer” approach to $2M and requires that a minimum size trade per proposal should be 30 ETH.

[Arhat]

Inherited cemetery. Find a buyer.

[rtp]

Hi DxDAO Community

I work with the Idle Finance DAO, and we have recently started similar discussions on the topic of treasury diversification. Excited to see your community exploring this as well!

Diversifying the Treasury balance definitely seems to be a direction many DAOs are going towards, which makes sense as it helps provide a solid runway to fund the protocol’s growth with higher certainty as it removes the market risk of being completely exposed to higher volatile assets. Your diversification reasons and allocation seem very well aligned to this thinking!

For those who are not aware of the Idle protocol, it is a yield aggregator seeking the best yield across multiple yield sources and assets. Currently, we are integrated with Aave (v1/v2), Compound, and DyDx; and we support large stables like USDC, DAI and USDT.

The Idle protocol is based on 3 pillars:

1. competitive optimisation technology (Dynamic Allocation Mechanism),
2. focus on security (5 audits, $500k bug bounty program),
3. and long track record (active since mid-2019, backed by Consensys).

The Best-Yield strategies maximises the returns while preserving high-security criteria.

For your reference: The average DAI APY has been 10.81% over the past 5 months; USDC has been 10.83%.

We have made a profitability calculator that you could use to better understand the yields you can get with Idle.

I believe the community is considering deploying a portion of your stablecoin assets into Aave and Compound, these are two key yield pools that Idle rebalances across for the Best Yield performance. So we would be more than happy to work with you closely if this is an option you would like to discuss further.

Best of luck with this!