[Informal Poll] Increasing the Surplus Buffer

I think some people are voting yes for risk related reasons and others are saying yes to increase the cash available for growth. Right now both groups are into growing the SB but I guess only one will be happy draining it for spending.


So we also need to get our arms around separating the actual SB from our operating cash. Because a year of expenses is (roughly) estimated at $32 million right now.

Suddenly the current SB looks very, very, very small.

Sooner or later (sooner!!!) we need to segregate funds used to secure the protocol against bad debt from our piggy bank. If we considered only 18 million DAI in the current 50 million to be legitimately SB, for instance, the risk to DAI’s backing is quite high even for very tiny ilks. (Hat tip to Brian and LFW for helping me better understand how backing works in a worst case scenario)


The method we use to calculate appropriate Surplus is to check each collateral type’s risk premium (can be found here https://simulation.blockanalitica.com/) and calculate the “portfolio weighted risk premium” which represents expected portfolio loss in percentage terms. The recent calculation we made, estimated the weighted risk premium of 2.5% and applying this to 2bn DAI outstanding debt from volatile collateral equals 50m DAI. This is in line with the current Surplus Buffer and we actually rarely (if ever) achieved that good risk coverage in the past.

However, this ratio won’t stay fixed going forward. If bull market kicks in or we increase DAI supply from volatile/risky collateral by organic growth (D3M, RWAs, stETH, G-UNI,…) we should be prepared to grow risk coverage in nominal terms as percentage wise if portfolio weighted risk premium increases. It also becomes easier to onboard risky assets, scale DAO and so on. From that perspective it is better to accumulate reserves in advance, because otherwise we’ll be lagging as we did many times in the past.

At the same time I also think we should start giving more focus on treasury management if consensus will be to grow the surplus buffer from 60m. At certain point only hoarding passive DAI in reserves becomes ineffective. And to be honest, probably the number one reason why MKR holders would prefer buy&burn is because they prefer their share of passive surplus money to be reinvested somewhere rather than held passively (in buy&burn scenario reinvestment is made in protocol from those who don’t sell MKR). If Maker is able make yield on at least part of Surplus, support for increasing surplus buffer could possibly be higher.

Also side note, I don’t expect 60m Buffer to be hit soon - unless rates increase or DAI supply from volatile collateral increases significantly to May levels, revenue will more or less equal costs and Maker would need to liquidate 77m DAI to accumulate another 10m in penalty fees to fill the remaining surplus buffer.


I agree with this. The only issue I see with the larger surplus buffer is that it is denominated in Dai, so we’re inadvertently pushing up the peg by holding Dai that doesn’t have any velocity. As this has been a persistent issue, I wonder if it would be worthwhile to explore saving rainy day funds in a basket of alternative assets so that we can free Dai up for the market. That said, even tens of millions of may not affect the peg much at this point anyways.


Voted yes to this. I would like to see a real assessment @Primoz from a what amount of cash do licensed banks need to have as I expect regulators would like to see Maker with that number or higher.

I want to echo a second sentiment regarding isolating operating cash from SB.

Agree with various comments @wouter @Aes and others startups really don’t buy back stock - they plow returns into growth.

I am also for buffer diversification and less concerned regarding PEG as per:

PEG is being driven by artificially high DAI demand due to LP farming. I see some moderate change of backing on DAI off of USDC to ETH predominantly. Not sure if this is due to SF or just a 2K-3K ETH price change as I have not looked at the numbers in DAI vs. percent.

I think it is important to realize that operational manpower costs now are over $30M/yr and SB basically has 1 year cash run ATM. Some analysis of what we can do with DAI surplus to roll into growing operational budget fund probably would be prudent. Realize it should be an easy matter to roll some of this back into the Surplus buffer in case of emergency if required so can still work as capital asset backing of the system in a pinch.

I think a real discussion as to a real targets and structures for:

  1. Surplus buffer (what is too much - why? )
  2. Operational expense fund (which can back up surplus buffer in a pinch. What is too little? What things can Maker invest in to get return and still retain good liquidity. What is too much.).
  3. What levels of 1, 2 before Maker would consider looking for buyers to flop MKR to pay operational expenses or deal with losses in the SB if we have operational funds we can fill the SB deficit with. Or what is too much and conditions to burn more MKR because MKR has a lot of cash on hand.

I think my main concern here is while we can talk about SB Maker would do well to also give some thought to when and how to line up MKR buyers (or whether to actually put up a DSR-B facility to borrow DAI with bond like terms to cover operational revenue deficiencies). Companies don’t usually just throw stock on the market if they need cash, they take time to find the best sale terms/prices, look at bond rates to borrow etc. This should all be factored into a cash reserve management strategy.


Don’t forget that MakerDAO can always mint MKR and it has 84K MKR catching dust. :slight_smile:

BTW—speaking of diluted token holders—Hellium just raised $111M. Raising funds is not a sin. I can we can say it’s part of “strategic patience.”

  • As discussed in the past, there is no way to invest the Surplus Buffer. The Surplus Buffer is on the equity section of MakerDAO not on the asset side. For most other protocols, the equity equals the treasury section which is on the asset side. Therefore, this one can be invested and managed.

  • Separating the surplus buffer from operational cash seems to be a good idea but it’s not really. That’s what I learned from my research for the Strategic Reserves. Some Core Units have contingency buffers that are out of reach of Maker Governance. It’s better for Governance attacks and it didn’t cost any PE time. Moreover, the risk of not being profitable (and not being able to generate cash flows to pay the teams) is somewhat uncorrelated to credit risk or market risk. So there is a diversification benefit of polling them.

  • I would refer to the [MakerDAO balance sheet] that RWF CU is maintaining (@Aes doing a great job at it). As you can see the Surplus Buffer is in blue, it’s not an asset. But the Operating Wallets are both as equity and as an asset. In that sense, you can say that there are some inefficiencies by having this cash unused. But it’s really a minor problem (4M DAI).
    (MakerDAO Financials - Google Sheets).

  • The equity requirement for banks is 3% of the balance sheet (let’s forget about more elaborated stuff like CET1 ratio, and it’s not saying anything that we are or should be in any way under this regulation). We are currently around 1% and we might have an issue catching up if there is hyper-growth life earlier this year. MICA regulation for stablecoin issuers will be 2% (not saying at all we are in the scope). Even for simple brokers (and we are kind of a broker providing margin loans on crypto), it is above 1%. So even before knowing how much exactly we need (which would depend on our risk appetite and our place in the competitive landscape), it is fair to say that it should be more than what we have currently to give us options in the future and avoid dead-ends.


Yes this a one of the bad post that everyone tries to avoid. But reality is that system is built in such a way that one day or other you have to hit that burn button wether you increase the limit or not. Its better that periodically that button is triggered as it causes fair valuation of makerDao. Not having a fair valuation would leave the system vulnerable to take over attacks ( Hostile takeover ) as MKR is not fairly valued and is at suppressed rates. Due to this I feel that increasing limit is not a sane option. I am open to hear counter arguments on the same.

The SB is an asset in the form of retained earnings. At the moment it also happens to approximate the equity of Maker because it has no other assets except the PSM (which is offset by corresponding liabilities). But it is cash that is contributing to equity, not equity in and of itself.

I believe you mean the capital requirement (CAR).

Please stop calling us a bank, broker, or anything else that has major regulatory requirements and is not us.

1 Like

Are we on PaperImperium Accepted Accounting Principles? :thinking: Retained earnings are not at all an asset, it’s on the right side of the balance sheet under equity. Assets are on the left.

Let me rephrase. The SB is the cash from retained earnings. My mistake for not clarifying more precisely when I said:

This is the original quote. I see how I may not have been super clear. But The SB is an asset. Which then increases equity.

…may I ask what your financial background is?

Maybe if you can provide a balance sheet, as I did, it would be easier to follow.

I have a college degree in accounting and did a fair amount of accounting design for an insurance company. Doesn’t mean I’m right or wrong here as it is cutting edge. But I think I have provided what is a framework that is compatible with accounting rules. And I fail to see how what you are saying could comply with those rules.


100% support. Also took me a while to realize but any Defi project paying dividends in this early stage, esp. w/o having a large treasury, is likely a blunder.


Could you post a link to how backing works in a worst case scenario? Thanks

I don’t have one at hand, but the surplus buffer absorbs bad debt. We haven’t seen any in a while I think, but that’s the first shock absorber, then more MKR is minted and sold.

1 Like

Voted to increase because cash on hand is needed for growth and/or stability in this phase, much more than burning. MKR price is appreciating as the protocols grows with the rest of the market anyway, so I don’t see the need for putting an extractive or price bump focus over cash on hand for expansion, stability, and unexpected/new expenses.


I have never been convinced that burning Maker will eventually be rewarded by the market. If ETH can drop to double digits, then rationality has completely gone out the window.


Respectfully, I think this comment is pretty misinformed. If you browse even just this forum, you can see myriads of work being done by the CUs on payroll, contractors, and active holders, all of whom receive compensation for their work. I don’t see how you could say with confidence that dai spent isn’t producing growth, even if you have an issue with specific expenses.

The protocol has been scaling pretty rapidly with more CUs coming on board, new integrations and connections being made by growth, new structures. If you check other posts there is still new hiring happening, and new CUs such as legal that some expect to be coming next. All of these require costs, and I’m sure there are a million other operational costs I’m not aware of. And it’s not like the protocol is exactly raking in the cash right now from what I can see.


Also I’m not sure what you mean by your comment about printing maker and creating hoards of cash. The surplus buffer plays an important role in the stability of the protocol. It’s not just idle cash doing nothing. I’m not sure if you think this is about the foundation MKR instead of the SB, but I don’t see how printing and diluting maker to cover operational costs instead of raising the SB and delaying a burn makes any sense.

thanks everybody for participating, i just opened the Signal Request for adjusting the Surplus Buffer.