Proposal Idea - Purchase ETH for Strategic Reserves [Poll]


Following the Oracle team’s gas budget request, I would like to discuss a proposal to purchase additional ETH held by the protocol as strategic reserves to hedge any sustained spikes in ETH price and Gwei.

I performed a simple linear extrapolation of the prior six weeks of actual average Gwei per day through Q1 2023 (directly below) and put in some different ETH price points to estimate the $ impact on our oracle costs (bottom table). To be clear - this extrapolation is not my forecast, there are a number of factors which put downward pressure on Gwei. The data is simply to illustrate how quickly costs can increase even with somewhat modest increases in ETH price and Gwei.

As you can see from the above, if ETH hits $15K by the end of Q1 and Gwei increases by 80% from 149 to 268 per day, we could see gas costs balloon by nearly 10x. We have seen multiple occasions over ETH’s history and even this past year of it’s price doubling within six weeks. As we’re all aware, crypto markets can move quite quickly and without warning. Having an additional backstop of ETH in reserves will provide us with a larger safety net and could save tens of millions of dollars of funds.

If sustained increases in Gwei or ETH do not materialize, the DAO can decide whether to sell a portion of the ETH or continue to hold in reserves.

To be clear, this proposal is supplementary to the funds the Oracles Core Unit requested for Q1 and for emergencies. If the Oracle gas budget is exhausted prior to the end of any quarter, the Oracle team can request ETH from the strategic reserve. This Strategic Reserve could also be accessed by the PE team for smart contract deployments and testing.


  • Hedges and mitigates higher oracle expenses due to sustained Gwei and ETH price appreciation
  • Provides price exposure to increases in ETH price (which as we’re aware has outperformed MKR)


  • Increases bad debt risk by having non-stable assets in the Surplus Buffer

Note - analysis assumes the same number of Oracles are active as were in October which will likely overestimate Oracle gas expenses as multiple collaterals are in the process of being offboarded.

How Much?

Regarding the amount to be purchased, our initial recommendation is for 2,000 ETH, currently valued at $9,000,000. This is a little over 7 months of gas supply at current ETH & Gwei, and 4 months of supply at $15K ETH and 268 Gwei. Would really like to get community input on this number and the proposal in general.

Longer term we will want to refine this number but today we are limited by the Surplus Buffer, which currently sits at $60M and will decrease by $8.8M once the Oracle gas budget is passed.

How to Implement?

I have already brought up the idea with @Derek and he has started discussing with the PE team. If the DAO decides we should pursue this path, I believe the team will be able to write up a requirement doc and technical implementation.

Should the DAO purchase ETH as Strategic Reserves to Hedge Gas Costs?
  • Yes
  • No
  • Abstain

0 voters

How much ETH (in DAI Terms) should be purchased for reserves?
  • 3M
  • 6M
  • 9M
  • 12M
  • None
  • Other
  • Abstain

0 voters


While this sort of event would represent a big increase in costs, I imagine we’d also have a comparable growth in profits because people are going to mint a lot more DAI. While the oracles will take up a larger fraction of the costs, I’m not sure if there is a risk to overall profitability. If anything, the protocol should make more profits despite higher oracle costs.

The issue I have with this is that there’s a decent chance we’re going to be buying ETH when it’s high and either selling low or just being stuck with it for a long time.

If we really do want to have strategic reserves, the price of MKR will be sufficiently correlated to ETH, even if it’s not perfectly correlated. It is possibly confirmation bias on my end but I also think MKR is a lot more undervalued at this point than ETH, if we want to try something like this. It still has the same risk of us buying high but at least it’s MKR that gets a nice price boost out of it.


I think it’s good idea - purchase ETH and rise Strategic Reserves level

Is there an item we could purchase that is effectively a call option on ETH price, i.e. an insurance hedge? I would support that. Holding ETH itself is speculation. Notwithstanding what we all know and believe about the Ethereum network.


I voted yes because I think we need to try and hedge somehow.

But this brings up a good point and if it is a better alternative to buy options, I would support that.


I think this is a very good point, and, on the oposite side, if eth prices drop this hedging stategy will result in a loss for the protocol in addition to the general decrease in profits we would probably experience in that case.


I would be more in favor of purchasing Uniswap DAI/ETH LP shares instead of ETH.


Can you do an analysis showing estimated change in profits to Maker if ETH say rises or falls over a 200% range.

If others posters are right, buying ETH now increases profit volatility…


My first question would be how this proposal relates to the surplus buffer? As far as I can tell this will directly remove funds from the surplus buffer and with it further postpone burning and worsening our ratio of surplus to non-stablecoin. 9M at current rates will take a month and a half to get back.

As for the need of this strategic reserve; In a scenario with rising ETH price we are most likely also increasing revenue and should thus be naturally hedged. In a scenario of rapid market downturn we are likely to see an increase in gwei cost but a reduction in ETH price. This is where we expects the numbers of liquidations to increase and if worst comes to worst have to dip into the surplus buffer.

On a political level this proposal seems to undermine the compromise that was reached in the big debate surrounding the surplus buffer. That is unfortunate and feeds into the narrative that Maker DAO doesn’t really have any intention of ever burning any MKR. The logical response to constant proposals to either rise the surplus buffer or drain from it would be to start a similar vein of proposals to reduce the surplus buffer and I think we can all agree that would be very unproductive.


I’m not aware of any trusted parties offering options on ETH and even if there were, the price of said options would be extremely high due to ETH’s volatility to the point where I believe purchasing outright would be the superior choice.

This is a fallacy IMO - should an airline not hedge their fuel costs if they expect an increase in air travel? The whole purpose of hedging is to minimize costs and thus maximize profits. What happens if Gwei continues rising but ETH price doesn’t and thus Oracle gas costs continue to rise and eat into profits?

There is always risks to hedging but there is an asymmetric risk reward to buying ETH as it can go up multiples, as can Gwei, whereas it can only decrease 100% in which case we are screwed anyways.

This is incorrect. Any asset purchased with surplus buffer funds will remain part of the surplus buffer. The only thing that changes is the volatility of the Surplus Buffer, which the community doesn’t seem to have a problem with given it’s desire to hold ENS and AAVE tokens accrued through D3M.

I spoke to Nik a while ago requesting an oracle gas cost estimate for the budgeting dashboard. At the time, estimated annualized oracle gas costs were only $1M. Oracle costs today are estimated nearly $16M/yr.

Let me be clear - if we had the foresight at that time to start buying ETH using the surplus buffer, we would’ve burned $15M more MKR than we have today (roughly 5,000 MKR tokens at current prices).


Here you go: ETH Options – inexpensive.
If my memory serves me right–I think I saw the March, or April $15,000 Calls at around $180, or $170, last month.

1 Like

After I wrote that I saw FTX offers them. Is FTX considered trusted?


I’m not sure I would consider these inexpensive - the IV is 6-10x that of the SPY. We also would take on significant liquidity risk, and counterparty risk by trusting a centralized third party.

Yet, I’m not aware of an airline that buys a bunch of jet fuel for the next several months or year as a hedge, they always do a forward supply contract, call options, or a combination thereof. Or any other business that relies on a volatile commodity. They are aware of that asymmetric risk – 100% down but multiples up – and don’t see buying the commodity itself for future use is worth the risk.


I believe there is a difference here because an airline will not necessarily have a natural increase in profits if fuel prices spike, but Makerdao will most likely see increased profits if ETH prices spike. As such there allready exist a natural hedging relationship. If eth prices increase, so does both gas cost (in DAI) and profits, if eth prices drops so does gas cost and profits. If we have an ETH reserve this natural hedging be less effective. Yes we will earn more if eth prices increase, but we will also loose more if they go down.


I wonder if that may be because of transport and storage costs



Can you please address some issues, as I think they are key in understanding what Maker would be hedging and the potential usefullness and risks of such a strategy?

  1. What is the Key Variable (KV) you want to hedge? Maker income?
  2. You are suggesting that we hedge KV that buy buying Eth
  3. What is the relationship between KV and Eth? How have you modeled this key relationship?
  4. How does gas pricing fit into this? How have you modeled this?

I believe we need a strong, consistent relationship between Eth and the Key Variable (KV) for such a hedging strategy to work well. Until one can demonstrate such a relationship, hedging can likely result in greater income variance…


I completely agree that long term we should use contracts or options to hedge oracle gas costs assuming scaling solutions don’t take care of that problem for us.

The issue is that crypto derivatives market is quite nascent compared to traditional finance and we would take on additional risks pursuing this strategy without an infrastructure in place that can confidently support it.

I also think your statement is a bit of an oversimplification, fuel costs for airlines have been flat if you compare Nov '14 vs Nov '21. Compare that to ETH which is up 630% over the past year, and 59000% over the past five. Not exactly an apples to apples comparison. There is also a fundamental reason that will likely drive ETH price higher - the merge - which is expected to reduce ETH issuance by 90%. Imagine what would happen to airline fuel prices if oil production was cut 90%.

I’m not sure I understand your argument, could you please clarify? ETH only needs to outperform DAI for this hedge to be effective, and Governance can decide how large of a hedge we want to make. If you believe DAI will outperform ETH, holding MKR is honestly not a good idea.

  1. Oracle Gas Expense
  2. Yes
  3. ETH is used directly to pay for Oracle Gas Expenses
  4. Yes, see snapshot below

A decision to hedge 0% of our Oracle Gas cost exposure is basically a bet that DAI will outperform ETH. I’m not in the business of timing market tops but I believe long term ETH will greatly outperform DAI.

1 Like

Thanks for the update.

  1. Can you now please analyze/model the Eth - Oracle Gas Expense (OGE) relationship? And show how strong and stable this relationship has been over time?

  2. Can you also analyze/model the relationship between Eth prices and Maker net income, as this is the ultimate variable Maker probably cares about?

  3. Can you demonstrate how a simple hedging strategy would have performed historically for both items #1 and #2 above?

I suggest that you look at several time frames - last 3 years, last 1 year and last 3 months

I do not beleive DAI will outperfom ETH in the long term, but hedging strategies are mainly used to hedge short term cash flows. If we buy ETHs using the surplus buffer at near ATH prices and there suddenly is a short term crash in the ETH price we will be in a situation were the surpluss buffer decreases in addition to a likely natural decrease in profits. Temporarely we can then find our self in a situation were we can be forced to sell some or all of the ETHs at a loss to be able to cover CU expenses.

I am not saying that this is a very likely scenario, but it is only 4 months since the total SF income was about the same level as current CU expenses. Since this is presented as a hedging strategy I believe the strategy should not result in an increased risk for the protocol for short term drops in the eth price. If options were used instead this risk would be eliminated, but then we would of course in practise pay a fee to eliminate the downside risk and also introduce a counterparty risk.