Should LidoDAO sell treasury ETH?

That’s where the 1.3m-1.5m$/month run rate comes from, or those are separated expenses? How long can the DAI cover current costs?

1 Like

Correct, conservatively the DAI alone represents about a year’s worth of runway just on DAI at current run-rates. However, will caution that there may be months with higher expenses, notably in the event of smart contract audits, which tend to be highly lumpy and very costly.

1 Like

Thank you for clarifying. Selling 10k ETH would cover around 2 years of expenses when added to the existing DAI, more than enough runaway, taking into account there’s a 5% staking fee going to the DAO and the possibility to cut expenses. I wouldn’t sell more than 10k ETH, and there’s a need to put in place a multi year framework to manage the treasury.


Hey everyone, I’m Nicole Maffeo the cofounder of SIZE Markets which is an on-chain OTC protocol. We chatted during the Dragonfly OTC deal over the Summer.

Agree with extending run rate although I am unsure if 10k or 20k makes sense. You could always start with selling 10k and then leave the door open to take a more active treasury management approach for the remaining 10K ETH over the next few quarters as markets play out. FWIW, this proposal pairs nicely with the recent stable diversification proposal.

Dependent on whether the proposal passes, I think that using SIZE could make a lot of sense. Would love to chat.



No, do not sell.

Instead, are there any non-critical cost buckets that we could cut to extend runway?

Another alternative is to collateralize the ETH or even wstETH if staked, which we could include as an option on the slate.

For eg

Other options are possible, for e.g. opening an unstaked ETH vault on Liquity, which has a lower liquidation ratio of 110% and no ongoing stability fees. You could essentially replicate the same vault as with Maker but with no ongoing interest fees, only the one-time minting fee of 0.5% iirc. The only disadvantage would be using unstaked ETH vs staked ETH for this but it’s certainly possible. We have not looked in great detail at variable rate CDP vaults just because of their unpredictability, though there are options such as Aave or Compound.

A reasonable slate of options could be:

  • Do not sell or collateralize
  • Sell 20k
  • Sell 10k
  • Collateralize 10k at 300%
  • Collateralize 20k at 300%

If either of the collateralize options wins, we wait for the stake ETH Snapshot to resolve:

  • If stake wins, we use the wstETH Maker vault
  • If keep unstaked wins, we use the Liquity LUSD vault

and whatever other options are proposed by the community as alternatives. If more options arise, we can run a final Snapshot to pick a CDP provider.

There are many good reasons to open a CDP position. It is probably our preferred option (stake 100%, CDP wstETH as needed), assuming no operational constraints. However, the operational reality is that maintaining the health of this vault may prove to be complicated for a decentralized community, particularly in periods of volatility.


In general I think it’s interesting to consider the idea of collateralizing part of the treasury (or eventually the DAO’s LDO holdings) for stables, but understand that it’s a lot more complex to manage something like this (e.g. would require people to closely monitor CDP health, and react via on-chain gov votes which would take 3 days to execute and will undoubtedly cause operational headaches).

I wonder if there are on-chain ways to do this (vs eg trusting a third party like Karpatkey to do this on the DAO’s behalf) and what the risk of doing so would be.

1 Like

I understand that there are financial engineering (e.g., collateralisation) options but to me the safest one would be to simply stake with Lido which currently (as of 21/2/23) gives a 5.2% yield.

Then, consider if we are running the DAO efficiently in the first place. Lido DAO makes a lot of money with the 10% reward fee. Why is this not enough to cover our operating costs?


Presently it is, at current prices. Realistically, the simplest operational set up would be to sell whatever protocol surplus is needed to finance operating expenses and fund it from that. It will depend on what token holders decide with respect to this proposal. At the moment we are not authorized by the DAO to monetize this surplus to cover operating expenses.

If no surplus is sold, operating expenses would have to be funded out of the treasury runway, which includes ETH. Presently there is enough of a runway for ~1-1.5 years at current burn rates, and may vary depending on actual expenses for things like audits or development times. If ETH is sold or collateralized, it would lengthen the runway somewhat.

Staking 100% of the ETH and selling none of it is a possible and very attractive option also. What these proposals are asking for are clear signals from token holders, who ultimately make the decision.


Did you take a look at DefiSaver for the on-chain management of collateralized positions? (

Managing any CDP position through on-chain governance is extremely complicated, and would likely require complicated infrastructure to set up automated stop-losses or keepers, and later to maintain. Using multisigs is out of the question for any of these proposals. These are the operational deadlocks for on-chain managed CDPs. If there are alternatives, we’d welcome hearing about them though.


Why not do both? Sell 10k and stake the remaining? Hedges outcome and gives community the best of both worlds. Can parlay the results into how best to manage treasury x extend runway going forward etc.

1 Like

It’s certainly possible! Vote your view! We are focused at the moment on making sure a maximum range of outcomes can be realized from the Snapshot polls.

1 Like

Hey Izzy and everyone! I am Santi from karpatkey.

I’d like to take this opportunity to introduce karpatkey to you and the entire community, given that you mentioned us in your post.

Thanks to state-of-the-art treasury management technology, karpatkey can leverage the Zodiac Roles Modifier toolkit developed by Gnosis Guild (who are also the creators of the SAFE multi-sig wallets) to offer a fully trustless, non-custodial solution.

Regarding your specific comment, if karpatkey were to collateralize part of the treasury following the DAO’s mandate, there are various risk management tools at hand to track the health of the positions. These include different types of alerts (CRs, prices, depegs, governance proposals, etc.) as well as automatic tools such as disassembler bots (EOAs in charge of disassembling a position when a preset threshold is met). There’s an entire risk management framework developed by karpatkey to make decisions and monitor the status of existing positions.

karpatkey is the leading DeFi-native DAO specialising in professional treasury development through extensive research and best-in-class tooling. With more than $350M of ncAUM (non-custodial assets under management), karpatkey has been selected by some of the most renowned DAOs such as Gnosis, ENS, Balancer and CoW to develop their treasuries in a non-custodial manner.

If you’re interested in learning more about the Zodiac Roles Modifier toolkit, you can find more information here and here. You can also find more information about karpatkey here.

1 Like

Hey Santi, thanks for sharing! This looks very cool!

How does this work exactly? First of all I’m assuming Safe is required (since this is based on Zodiac which is an extension to Safe), so that also requires moving the assets from the DAO treasury to the Safe.

Who is actually controlling the “Safe” in this case, I guess karpatkey (and/or in conjunction with someone authorized from the DAO)? In the examples outlined in your tweets, it looks like there is an allow-listed set of “operators” of the Safe (kinda like traditional companies might have someone (or a combination of people) in a Treasury department who can execute things in their bank accounts / trading platforms etc), but I think that’s one of the things that (at least based on the discussion in this and adjacent threads) contributors have thus far suggested to avoid (mostly due to lack of resources/time/operational capacity to effectively manage and properly implement something like this).

My pleasure, Izzy. :slightly_smiling_face:

Yes, a Safe is required for this process. However, transferring funds from the DAO treasury to a new Safe is not mandatory since the Manager Role can be created by an existing Safe. Nevertheless, it is recommended to create a new Safe controlled by the DAO to separate the funds managed by kpk from those that are not.

It is important to note that the funds are always under the sole control of the DAO. If karpatkey partners with the Lido DAO, the latter would deploy the Zodiac Roles Modifier contract on the wallet holding the funds and then grant kpk manager access. This means that kpk would be able to take specific actions on behalf of Lido without any possibility of doing anything different than what’s been permitted. It’s an innovative and secure technology.

The Lido DAO controls the Safe. Karpatkey would only control the Safe that has been granted the Manager Role. There is no need to rely on any specific resource or operation capacity provided by the DAO.

1 Like

Thanks for the additional info! Just one point to clarify here:

The DAO Treasury is not a Safe, it’s an Aragon vault. Due to that, I don’t see how this could work without transferring it to a Safe to begin with.

There may not be an explicit need (i.e. if authority is given to karpatkey to do all the actions) but I’m not sure that’s the most reasonable/prudent approach from the get-go.

1 Like

Thanks for your comments.

You’re right. Then a transfer to a SAFE is a required step.

karpatkey would only be able to operate within the boundaries established by the DAO. I believe that DAOs in general should focus on their core business and leave treasury management to dedicated parties rather than trying to tackle everything themselves.

We’d like to take this opportunity to chime in and share our view on this proposal.

According to Aragon’s records, the DAO’s treasury exposure to ETH at the time of writing is approximately 72% of the entire portfolio. Lido should diversify the portfolio and secure enough stablecoins to guarantee a minimum runway of 2 years’ worth of expenses—without sacrificing the ETH price upside once the bull market returns. Current run-rate’s range mean for the DAO according to this proposal’s context info is around $1.5M/month, which accounts for roughly $33.6M over the new 2 years.

One potential path is to sell 10,000 ETH (~$16M) and use the existing DAI holdings (~$17.35M) to fund the operation over this period. The proposed transaction could be executed through CoW Protocol and their Milkman project, which allows to defer the choice of the limit price from order initiation (pre DAO vote) to order execution time (solver settling the trade). ENS has already executed an identical proposal successfully.

Additionally, staking the remaining ~10,304 ETH would increase capital efficiency without adding operational complexity to the DAO, while promoting the protocol’s own value proposition.

While there are more complex alternatives such as collateralizing stETH to borrow DAI on MakerDAO, the DAO may not be able to actively monitor these positions at this stage.

Good treasury management practices are critical for any DAO looking to survive in the long term, and it’s encouraging to see Lido moving in this direction.


how are we back to this place? didnt we just sell a bunch of ldo tokens to do the run way?

why are we here again