I agree that the denomination of Lido, the protocol, should be in ETH as that’s what the smart contracts require. But I don’t see why that needs to translate to the operation of the DAO. While the incoming assets would be in ETH or stETH (5% of yield earned goes to the DAO AFAIK), all of the expenses are denominated in USD or equivalents. So, when we think about runway, which is the most important component of the DAO accounting, we would be looking at USD outlay per year, divided by USD owned by the DAO.
I think it’s actually very important that the accounting denomination be set correctly, because it governs so much. For example, ETH and stETH should both be seen as investments. I don’t think you can see ETH as a currency for the DAO, when none of the expenses are denominated in ETH. The only reason to hold it is because you hope that the price will go up in dollar terms, such that you can pay for a longer runway, in dollar terms.
That seems pretty clearly the case, when one of the proposals is “should LidoDAO sell its treasury ETH?”.
The above all presumes that the treasury today exists to further the goals of the protocol by paying grants, contracts, contributions etc (all denominated in LDO and stables), and should NOT be utilized for any other reason.
Let me give an example so that it’s more clear what I’m trying to express.
Currently we have $20.1m of stables. If we only intend to hold stables and NOT to invest them into yield generating opportunities, maybe a reasonable split would be something like:
$15m DAI
$5m USDC
This is based on having a $60m treasury, so $15m DAI would be 25% of the treasury and maybe that’s a reasonable maximum for how much of the DAO Treasury a single token should make up (just hypothetical… roll with the example!)
Now, let’s say that we decided that 50% of the $20.1m is going to be invested into treasuries. Now, we only have $10.05m of stablecoins to split up. In which case, you might argue that you can hold 100% of the $10.05m in DAI, because it’s well under the $15m (25%) limit that you come up with previously.
Hopefully this shows how the utilization of currency makes a meaningful difference to how to diversify. Primarily because diversification is not necessary, or ideal, when the raw amount becomes smaller.
FWIW, generally my opinion here is that if the
Expenses is $18m / yr (based on numbers from here)
Income is $20.3m / yr
The DAO is already profitable, but getting to a position where the yield / interest on assets can pay for expenses would be powerful. This is critical because we currently face basis risk whereby expenses are in USD and income is in ETH, meaning that we could face deficits and hence have a limited runway of operations due to ETH price movements.
To get to a point where our asset yield covers these expenses, we would need about 400 million in assets. So, it’s not going to happen in the short or medium-term. But, moving there as fast as possible (by growing the treasury in $ terms) will reduce the impacts of ETH price volatility by minimizing the % impact of reduced USD income from Lido fees, compared to fixed expenses.
With $60m in assets, you can likely earn around 4.5% / yr on active capital. That would generate $2.7m / yr, which is more than the excess profit that the DAO is currently operating on ($2.3m). So, deploying the assets would double our yearly surplus. With the power of compound interest, that will allow us to grow our runway far more than 2x as fast as without investing the assets.
Worth quickly touching on why basis risk matters:
If ETH price drops 50%, which could happen, that would take us down to a price we’ve already seen in the last 6-months. All the income is denominated in ETH, and so our USD income is cut in half.
So rather than having $18m in expenses and $20.3m in revenue, meaning a $2.3m surplus, we would only have $10.15m in revenue, meaning a ~$8m deficit. That 50% price drop would bring our treasury down to ~$40m, meaning that we’d have a 5-year runway before the DAO has to halt operations, under a 50% price drop scenario.
Even if we look at a horrible 80% price drop (maybe it’s not exactly 80%, but imagine large slashing event causes Lido outflows, combined with ETH price drop, creates some sort of scenario where numbers below look ballpark for the event):
Income goes to $4m, expenses at $18m, treasury is worth $28m now, deficit would be $14m per year. That still gives us a 2-year runway, even under this horrible 80% ETH price drop scenario.
Look, this is all napkin math, I haven’t actually modelled this out, but it’s fairly obvious that unless something goes horribly horribly wrong, the treasury is going to be able to continue to operate as is for at least a couple of years. If it can make some efforts to cut expenses during down periods, even better.
So, I don’t think we need to be uber cautious here. We don’t need to start selling ETH to DAI. But, we really should start utilizing these assets to generate yield. If we do see sustained ETH price drops, taking us into a deficit, then we will of course need to either start cutting expenses or selling assets to fund expenses. The first of which to be sold should be ETH / stETH, to reduce basis risk.
For a more immediate structure with the existing assets I’d propose the following:
- Stake (or buy) $30m of stETH and earn 4.4%+.
- We now hold $20.1m DAI, $39.2m stETH.
- Buy $5m of OUSG (Ondo) using DAI, earning ~4.5% APY
- Buy $10m of 0-5 yr treasuries (Backed Finance GMBH) using DAI, earning 3-4.5% APY
- We now hold $5m DAI for short-term cash, $39.2m stETH, $15m treasuries/bonds
Then, all the income coming to the DAO due to the 5% fee in Lido should be automatically sold to DAI and be used to pay expenses. All excess at the end of the month, beyond a minimum $5m in petty cash, can be re-allocated to purchase treasuries / bonds. Such that, our exposure to ETH / stETH will reduce overtime (as a %), giving us less volatility and more security.
Final comment here is that after conversations with people involved with this, I think it’s super important that we codify and are explicitly that the DAO Treasury is to be used for runway, grants etc. It should NOT be used as a 2nd insurance fund in the case of slashing, it should not be made to make users whole in the case of some loss event. The size of the treasury is so small compared to any potential large event that it will make little difference to recovering losses, yet will wipe out the treasury entirely. It’s critical that the treasury can plan going forward, which means that it must be ring-fenced and used only for explicit use cases.
If the community is concerned about slashing losses, buy an insurance policy or vote to direct more funds to the insurance fund. The treasury should not be accessed for these cases, otherwise you risk the entire protocol itself.