Discussion: Investing the Surplus Buffer


while MKR is mooning, the “strategic patience” people in MakerDAO’s chat are rationally talking about the Surplus buffer.

Some discussion is already going on in separate threads regarding the need to increase the SB to keep up with the larger amount of DAI minted (now >1.3 billions) and with the need (soon, when the Foundation will dissolve) to self-sustain the DAO (read: salaries).

Today @NikKunkel wrote in the chat:

plus we could always reinvest funds in the surplus buffer to earn yield… since theyre not user funds but protocol funds

Which makes a lot of sense. @LongForWisdom wisely replied with:

  • So having them in compound or Aave sounds bad, because it directly makes our competitors cheaper. Having the buffer + PSM funds as DAI-USDC uniswap tokens seems nice.

I started this thread to collect good ideas along the lines of: how can we invest the (inactive part of the) SB?

I start with my 2 cents.

IDEA: As a variant of what @LongForWisdom proposed, it might be cool to LP on Uniswap the MKR/DAI pair. So, concretely:

  1. Use some of the money of the SB to buy MKR (this is separate from the burning mechanism).
  2. We then become Liquidity Providers on Uniswap on DAI/MKR.


  1. MKR gains liquidity, of course, which is always a good thing.
  2. Unlike other forms of investment, which are questionable (everybody could say: nah, just give me my share and I will invest it by myself!), this investment does the good of all MKR holders, increasing liquidity.
  3. Since MKR holders’ goal is to vote to increase the value of MakerDAO (and therefore of their MKR), all MKR-holders should be considered as bulls on the MKR/DAI pair, hence LP this should make sense for all MKR holders.
  4. Finally, of course, we get the LP-fees, producing the desired yield.
  5. Additionally, if the MKR token goes up in value, the LP tokens gains in value (vs the initial investment measured in DAI) and so we get even more gains.


  1. What could go wrong? The MKR token goes down in value. This means that our LP tokens loses value (bad news) but it also means that it represents a growing and growing amount of MKR (>50%) vs DAI (<50%). Well, if the loss is judged too heavy, we can return the LP-tokens, collect the MKR and burn them (feeding the burning contract instead of buying MKR in the market) so overall it’s not a terrible deal.

Anyway, this was just a random idea :slight_smile:
@LongForWisdom’s idea is much simpler and basically risk free, so arguably better.
I am curious to see hear other ideas.


Yeah basically we could alter the flap to basically buy MKR and then pair the next 10K DAI and put it on uniswap. I think it is a fabulous idea actually because then we earn a return on the surplus while providing liquidity. MKR goes down in price we end up with more MKR and basically can opt to burn it. (we end up paying less than just straight up buying and burning it) and we earn some return to boot.

I also thought supplying DAI/USDC was a good thing suggesting uniswap but curve may be more appropriate since these are both going to be pegged 1:1 there is little risk of IL loss though now with PSM in force this may not get much volume. The key point here was to make sure this liquidity doesn’t get locked up.

I really like the idea of doing this with the DAI that is going through flap. flap 1, buy MKR, take second flap and pair the DAI with MKR in a LP contract. buy and burn like 3(or N times) more MKR, then do the buy and pair into uniswap again. Change N based on the MKR price in a clever way and one gets an element of MKR price management as well. :slight_smile:


my two cents this idea or some version of it likely makes a ton of sense, but this seems squarely in the domain of what i hope the strategic reserve fund turns out to be.

Personally i think that is where we should be focusing our efforts atm in order to get that off the ground.

1 Like

I support the use of SB funds for stable investment and can use a variety of combination methods for investment management.

Love this idea and hope everyone else sees the merit in having a dynamic stability buffer.

Great initiative, two things for which I’m against:

  1. Providing liquidity is one of the few profitable activities mkr hodlers have currently with the token. Providing further liquidity would dilute their revenue stream

  2. Even though it may seem that it is more profitable it so solely as paper gains since the moment we use the funds to cash in it could be signalling a sell sign to the markets if we the community are cashing in on the tokens than rather using the dai reserve

I would on the other hand favor a eth-dai pool with eth as a SoV asset to later in time have nodes in Eth 2.0 from which the revenue could be used to fund other activities

ETH-DAI pool and MKR-DAI pool, we can conduct small-scale experiments at the same time, and we can have multiple combinations at the same time.

1 Like

I am all on board using DAI pairs to provide liquidity and have the stability buffer generating passive revenue.

I fully agree with @iammeeoh

A few points:

  • There is no usable DAI in the surplus buffer, it only contains accrued interest from vaults which isn’t a currency.
  • The strategic reserves is not the best way to achieve anything here. It might have a small role to play in the case of ES but we can hardly call it a priority. I mean, that works but it doesn’t leverage the main power of Maker: DAI.
  • Balance sheet manipulation is now strictly a better way. Significant work remains to be done and my hobby time is not enough for that.
  • Uniswap DAI/USDC is probably the best option to start with.

I really want to highlight that using smart vaults (or whatever the name should be for those active automated use of our balance sheet) is the most promising aspect I can see for Maker in 2021. Just like the PSM was the solution to solve the peg.

1 Like

I really like the idea and this makes a lot of sense to me.

Probably we should simply onboard in PSM assets which accrue interest (ie. aDAI, aUSDC). That way we have to do nothing “technically difficult” (only adding another asset to the PSM) and that kind of asset will be “invested” by themselves.

With some modifications, this could apply to a big family of assets (cDAI, fDAI from Harvest, yCRV), the price of this type of assets can be always queried onchain.

Same could apply for the creation of “auto-invested reserves”. We’ll not have to handle with voting strategies and making a Yearn.Finance ourselves for our assets; in this case, this type of assets have this implemented already within themselves.


I think this is a great idea, if we do this ourselves don’t need to be discussing so much about using other protocols (which compete with Maker).

Yes but these people can and do add/remove liquidity where as governance generally would mostly add and only remove tiny portions as returns accumulated (or simply leave it there permanently). So yes I agree with your (2):

The biggest thought here is that the protocol could lose DAI here in a market crash even though we would accumulate MKR. I think in the loosest sense doing this is probably the best way to stablize MKR prices both on the up and downside.

This part I don’t understand because basically if the ETH price crashes we will be selling ETH and the above won’t act as a reserve per-se. Also the ETH-DAI liqudity pool is one of the largest.

In the end looking at comments I think I have to agree with the most cogent and relevant post on this issue:

We should begin to look at what the goals should be of such a strategic reserve fund. In my own work previously I believed such a fund should act as:

  1. Fall back for DAI liquidity in the case of a negative surplus (fill a surplus gap before flopping MKR). Hence I see such a fund as seeking not just return but also retaining or increasing DAI value by putting assets into protocols that were 100% liquid. (which is an argument for why we shouldn’t use ETH-DAI or MKR-DAI btw)
  2. To use this fund to build a return for the protocol which then would be used to burn MKR.

Regarding any ideas of what to do with such a fund perhaps we should focus on the goals for such a fund first in a community sense as these goals are going to drive what we invest in.

1 Like

Makes sense, as long as governance does not commit to actively sell mkr

If eth were to crash so will the rest of the erc20, ethereum is currently the platform were maker operates so we already have the platform risk anyhow and eth is more monetizable than mkr, additionally the eth-dai pool currently yields 74%, can’t seem to find the mkr-dai pool on uni, we could eventually sell the profits for eth and stake it in the eth 2.0 contract. Basically what I mean is we are already subject to ethereum risk, might as well make profit out of it.

Anyhow @SebVentures already pointed out in respect to surplus buffer usage so I have pending to look in depth to his balance sheet manipulation write up

Yeah I forgot about ETH-DAI having a pretty high LP return at 74%. I have not looked at the MKR-DAI one.

I think we need to settle on a goal for a surplus reserve. Since it is called the surplus reserve vs. the surplus ‘investment’ we probably want to focus on capital preservation while getting ‘some’ return.

Personally I think a surplus reserve should have as a key component DAI value preservation, high if not total liquidity (when we need it we want access and to not screw up markets or worse having it impaired somehow), with return generation being important but distinctly less important than capital preservation.

I appreciate the discussion and points @mario