Move all idle DAI to sit in the Dai Savings Rate contract until needed

Background

Part of the Finance roadmap is to design a long-term plan to manage DAO resources prudently and in line with the DAO’s objectives. As illustrated during some of the discussions that took place during Lido’s Diversification Rounds, managing the DAO treasury needs a structured approach that solves for lengthening the DAO’s runway until it is self-sufficient. Key to this structured approach is a framework that all token holders can get behind, which will take some time to iterate through community discussion and governance.

However, until that time comes, there are easy-win actions the DAO can take without putting the protocol at significant incremental risk relative to its situation today.

Proposal: All idle DAI must sit in the Dai Savings Rate contract until it is needed and withdrawn

There are currently (at press) ca. 22m DAI in various governance contracts that belong to the DAO or are deployed to operating multisigs at various operating entities contracting for the DAO:

Aragon 19.06m
LEGO 0.06m
RCC 0.71m
Other operating wallets 2.12m
Total 21.95m

This proposal is very simple:

  • For Aragon balances:
    • Aragon vote to move all treasury DAI to the Dai Savings Rate contract, other than the amount needed for the next EasyTrack funding round
    • Until EasyTrack infrastructure can support exit and transfer in the same step, any future DAI withdrawal Aragon votes will therefore be composed of two calls:
      • exit from DSR through an Aragon vote
      • transfer to wallet through regular EasyTrack without an Aragon vote
    • A final LDO vote could leave some amount of DAI as working capital to minimize unnecessary governance interactions
  • For LEGO, RCC and all other operating wallet balances:
    • This proposal provides delegated authority to multi-sig signers to execute deposits and withdrawals from the Pot on a discretionary basis, with the aim to maximize the amount of Dai exposed to the DSR
    • Whenever outbound transfers are required, the necessary Dai will be withdrawn from the Pot by multi-sig signers first

At the proposed 1% DSR rate, this could represent up to 0.2m in additional revenue for the DAO at low incremental risk, depending on our cash burn rate and the amount locked into the Pot.

Long-term work: Build EasyTrack infrastructure for multi-step function calls

Currently, there is a cap on how much Dai could be exposed to the DSR given the (intentional) limitations of the EasyTrack process.

As part of the broader work to manage Lido DAO resources against the DAO’s needs, the on-chain operations workstream will design and plan a new version of the EasyTrack contracts that could facilitate exit from the Pot followed by a transfer call, allowing for 100% of the DAO’s DAI balances to sit in the DSR at all times.

Risks

The Pot contract is an integral part of the MakerDAO system and pays out a savings rate from accumulated stability fees.

The risks involved in depositing DAI into the Pot are incurred when the overall stability of Dai is compromised, for e.g., if the MakerDAO surplus buffer is emptied, or if Dai were to become collateralized by bad debt. In this instance, whether the Dai is in the Pot or not, would represent the same amount of counterparty risk, other than needing one more function call to move away from Dai to any other token through a swap.

Implementation

Implementation will be reserved at the discretion of the on-chain operations team provided there is no incremental risk, which may include using the CHAI contract, an ERC-20 implementation:

10 Likes

Thanks for the proposal!
The idea looks reasonable. The only thing I’m concerned about is necessity of the Aragon vote to withdraw DAI and make it available for the EasyTrack. Main risk here is that we will basically loose all agility that EasyTrack allows for the trusted payments.

So if Lido want to keep most of the DAI in the Dai Savings Rate contract than EasyTrack able to withdraw funds from it is crucial. For simplicity it can be a standalone EasyTrack just for the withdraw purpose.

2 Likes

Tend to agree–EasyTrack implementation is the right solution for this. Fortunately there are ERC-20 adapters that can make the EasyTrack processes easier to implement.

The potential way around this is to still hold some DAI in treasury + use wrapper token for the position in the pot — this can be withdrawn from the treasury with EasyTrack.
Would note that for that plan one would have to get new EasyTrack motion factories (basically everywhere where DAI is used now)

It seems reasonable to have idle capital working for the DAO, however I share the concerns listed previously and tend to back @kadmil 's proposal to have portion of the funds outside the savings account for agility.

Also would be great to understand risk:reward ratio?

Subjective opinion from a business perspective is that we should have another chunk of “stability fund” for operations outside of any contracts to preserve operational capacity if something goes wrong.

I support this proposal and I think it is a great way to increase the runway with minimal risk added.
Dai Stablecoin System has been audited and battle tested in production, as well as CHAI.

What would be the expected timeline for building production ready easytrack infra to relieve the operational effort?

Checking with @kadmil, assuming there is a single-step EasyTrack motion that could, in one step,

  1. Exit Chai
  2. Transfer to an operating multi-sig

Would this not effectively allow 100% of the DAI to be locked in CHAI at any given time?

Re: risk-reward, the risk is fundamentally equivalent to holding Dai. The Dai simply receives a portion of the Maker Surplus Buffer at the DSR rate every second.

Not riskless by any means (holding Dai or any other token in itself carries some risk of course), but there is no additional credit risk layered on as a result of the Dai being, for e.g. lent out or deposited in an LP pair.

4 Likes

In that case makes complete sense to do so. :+1:

So, from the ops perspective, I don’t think that EasyTrack motions for wrapping / unwrapping are good way to go, tbh.
If one can just hold Chai (need to check the contract for more to be 100% sure it’s safe to just use the wrapper token), the easiest way would be to hold chai in the treasury (wrap funds through Aragon) + have “chai payment processors” in place of “dai payment processors” for funding all the committees.
Must also note that deploying all the infra required isn’t trivial (more "checking we’ve deployed the right things & configured those correctly than only deployment, we’re quite ready in that regard).

1 Like

Another thing to note, I’d argue it’s worth leaving some portion in DAI anyways for any contingencies + nearest ops expenses

2 Likes