[DRAFT] DXyield

With the rising price of crypto over the past few months, DXdao has found itself holding a significant amount of assets. While we’re working toward more transparency around the DXdao treasury, we also have to think about allocating our capital efficiently - rather than just having it sit in our treasury. With the surge of DeFi protocols over the past year, there are a number of opportunities for DXdao to yield on its assets. The Treasury Diversification Proposal V2 authorizes DXdao to yield on a subset of its assets.

DXyield is essentially a multi-objective optimization problem - where we want to maximize yields, safety and retain diversification. Yields are easy to quantify, safety of protocols and diversification are somewhat more subjective. Total funds in a protocol can serve as an indication to the protocol’s safety.

Authorized funds:

  • 5M USD Stablecoins
  • 1000 ETH

Authorized Protocols:

  • Compound
  • Aave
  • Yearn

The treasury currently holds the following stablecoins:

  • $1.8M DAI
  • $2.1M USDC
  • $0.9M sUSD
  • $0.8M USDT

with another $2.1M already authorized to be converted into stables. Which will result in a total of ~$8M in stablecoins (this excludes $1.4M provided as LP to Swapr).

Compound

Compound is one of the more established DeFi protocols, currently holding $12.5B in assets.

30 day average APY:

DAI USDC sUSD USDT
3.85% 4.73% n.a. 5.56%

Aave

Aave currently holds around $16.2B in assets.

30 day average APY:

DAI USDC sUSD USDT
6.27% 8.49% 32.55% 11.76%

yearn.finance

yearn vaults exist for all stable coins held by DXdao.

Net APY (as displayed on yearn.finance):

DAI USDC sUSD USDT
11.88% 14.2% (new vault) 15.39%

Proposed execution plan

The following execution plan was devised with the following assumptions:

  • Diversification - we don’t want to deposit all our capital into one protocol.
  • Safety - we are more comfortable with depositing larger amounts into established protocols.
  • Yields - we want to maximise yields while respecting the two points above.

The following proposal allocates the $5M funds as follows: $2M in Compound, $2M in Aave, $1M in yearn.

Nonce Protocol Stablecoin Amount APY 30D AVG
0 Compound DAI 5k* 3.85%
1 Compound DAI 995k 3.85%
2 Aave sUSD 5k* 32.55%
3 Aave sUSD 995k 32.55%
4 yearn USDC 5k* 14.2%
5 yearn USDC 495k 14.2%
6 Compound DAI 0.5M 3.85%
7 Aave USDT 0.5M 11.76%
8 yearn DAI 0.5M 11.88%
9 Compound USDC 0.5M 4.73%
10 yearn USDT 0.5M 15.39%
11 yearn ETH 100 4.8%

*as protocols are used for the first time, we will use a test transaction to ensure everything works correctly.

Assuming rates and all the rest stay equal, the allocation above with the current 30D AVG APY rates should yield ~$56k a month (excluding the 100ETH in yearn). Which already covers a significant portion of our monthly burn rate.

The execution plan above requires, and thus depends on, the Multicall installation on mainnet and a relayer to interact directly with the yielding protocols.

Funds allocated to DXyield should be allocated with a long-term position in mind. Whereby, we should attempt to minimize withdrawals from our positions to not only avoid network costs, but mainly to compound our positions.

If anyone in the community has suggestions for DXdao yielding and/or any feedback, please feel free to share them in this thread or drop me a DM on keybase. As a general rule protocols should be open-source, decentralized, and well audited. If we find reputable new DeFi protocols, we can include those in future Treasury Proposals.

4 Likes

I believe it is worth also giving Idle.Finance a thought for relieving operations and management of future allocations down the line.

Without having to manually switch funds between lending protocols, it will introduce automated asset allocation and aggregation for our interest-bearing tokens. We can thus optimize profitability and achieve better yield with less friction.

DAI 5.256% | USDC 7.223% | USDT 9.966% | WETH 1.546%

Idle received an audit by Quantstamp on April 26th 2021, which found no high risks and any low or medium ones have been fully resolved.

It currently has +$200m TVL.

Yearn is difficult in the regard that you have their compound risk which they redeploy to different strategies. Most stables are added to a curve pool and farm crv in the background.

So if USDT fails you end up with holding worthless USDT, even if you deploy dai on yearn. (yearn uses your dai in the background to add it to a pool USDT/DAI/USDC on curve)

I, personally use yearn, I have quite some money here. But I think we should look that we don’t end up all the stable in one bag.

But if you take aDai you get from aave, put it into aDai/aUSDC/aUSDT pool on curve, then take this token crvAave to yearn you have a similar risk profile, but at least know where your risk is.

Right now yield is then the aDai yield 5.4% combinde with the 8.92% from crvAave.

1 Like