[Signal Request] Increasing the Surplus Buffer (2021-11)

While there may be bias on behalf of the author, I don’t think this is reflected in the choice to make the implied yes the ‘anti-burn’ option. Ideally the implied ‘yes’ should always be the change from the current state of affairs, which I strongly suspect is how @ultraschuppi approached this.

If we leave things alone, do nothing, make no change, then we end up burning.

If we change things, do something, make a change to the parameter, then we end up not burning (or burning less.)

It makes sense to me for the ‘yes’ option to always be the option that advocates for a positive action, and for the ‘no’ option to be the current status-quo. I think to have the opposite convention, or to have a convention where this flips randomly, would be confusing and generally a bad time.

That said, all the points you made about trying to solve this in a more durable fashion make a lot of sense to me. It would be great if we could determine a policy for burn versus buffer such that we don’t have to vote and rehash the same arguments every few months.


I agree and I dont see how Schuppi could have phrased this signal requests differently, and dont think he should since he seems to be leaning towards less burn, more cash. IMO it is actually fine that there is always a bit of bias in a signal request (although it should of course be minimized as much as possible, which I also think it is here), because they are all about individuals bringing forth ideas about taking initiative and taking action. In the end MKR holders will have the final say no matter what so it’s not a failure if something contentious goes through the SR process.

My point is that given how close this vote is regarding partial burn vs no/minimal burn, it might be advantageous for the pro-burners to try with a different signal request that can somewhat counter the worst negative outcome from this signal request (not burning at very low prices) - specifically by linking price to how surplus buffer increases are implemented

Side comment: This is always how I believe burn should be done, buy low sell high and all that - an automated approach to valuation model-linked burning is described in the Ember design in my original Sagittarius tokenomics post, and I’ll post an even more advanced and powerful version of it soon in my v2 tokenomics documents.


last time I ran a SR about the Surplus Buffer the SR passed but the corresponding onchain poll failed. MKR holders do pay attention - that’s great!

For me it is more about people making decisions about relevant topics than to get my opinion implemented here.


There was a point where using a similar auction system to liquidations 2.0 was being mooted for the surplus auctions. Is this still in the pipeline or has it been shelved for now? Only asking because it might help increase participation if that’s a point for concern.


I’m wary of a culture where individuals do work in public, only to have it be rejected without nuanced feedback. I sense an impending burnout from contributors who are active in governance but ultimately didn’t have the opportunity to acquire a financially or politically impactful bag for themselves. This is not singularly related to the above quote, it’s something I think we’ve seen a fair amount of in the last few months.

As was mentioned in the facilitator off-boarding post, there’s no way to force anyone to continue to work on the project through voting.

Is anyone else thinking about this? Feel free to DM me on discord (same handle), I’d like to help prevent, for lack of a better word, brain-drain.

crossposting for visibility so we get this sorted out once and forever - never gonna happen, we gonna fight on different parameters later :wink:

1 Like


Lets adopt a sensible programmatic approach and then can argue about how to fine tune it later

Now we kind of look like a dog chasing its tail around and around


we should do an approach that brings all elements together. A burn to get more traction for the MKR token. The buffer to bear safety and security for the system. And also think about expanding the system into the future. We should harvest the fees while they last and build out other streams of income. We should not try to solve each issue when they come. we should predict the issue at hand and proactivity. solve them ahead of time.

1 Like

How about burn 50% of the net income and put the rest into the CS until we reach a certain percentage? Risk recommends 2-3% of risk weighted assets, right?

Then save 50% of net income until we reach 2.5% (mid point). Once we reach 2.5%, say burn 75%?

But I also stongly recommend Maker get more comfortable with Risk’s methodology and recommended levels…

But I do think it makes sense to manage to a Risk provided number. And then we can argue over the Risk Method/Assumptions/Number. Which would be progress as Maker would be focused on how to measure and manage risk

1 Like

Why don’t we just let MKR to be burned.


Because we’d have to mint and sell MKRs in case we need more and that has a (very) negative impact on price too; that’s much worst than not to burn at all.


Then let’s set surplus buffer to INFINITY, wait for the worst, and see what happens with MKR price… This is negative way of thinking and if we are really thinking that tomorrow BTC price falls to zero. What are we doing here?

Burning is the only factual event causing MKR price to rise. So let’s be conservatively optimistic about the market and let MKR to be burned in therapeutic quantities. If fair MKR price is 5000-6000 USD then let it to be burned at 2800-3000 USD.

1 Like

I don’t think we need to set surplus to infinity :slight_smile: but having at least a few % saved in capital against non stablecoin debt is prudent and will likely end up with better MKR performance and more tokens burned over the course of a full business cycle.

On Black Thursday last year, the $6 million in losses cancelled out over $10 million worth of accumulated earnings (due to buying MKR at relatively high price with burns (~$500), and then needing to sell at low price (~$275) to recapitalize the system). If maker had held a larger surplus at the time, some or all of this extra $4 million in effective losses could have been avoided.


A quick add-on to Monet’s well-made point – if the SB had covered all of the protocol losses from BT, we would have avoided the serious reputational hit that we took as a result of what transpired + the corresponding burn.

FWIW, I think we should look at another, similar situation where protocol losses lead to a depleted SB and then to flops as a “never again” sort of event. A big SB not only insulates us from going under water during a Black Swan but optically communicates to the outside world, and those who are keen to regulate our space, that we take these security and consumer protection issues seriously. And protecting the health of the protocol, given its TVL and widespread usage, is a consumer protection issue, imo.


The Surplus Buffer doesn’t prevent a Vault from being liquidated for 0 DAI in a Collateral Auction. It just prevents the Maker Protocol from having to mint new MKR to cover debts that aren’t covered by the Collateral Auctions.


Thank you and yes, I know that.

Yet my point still stands: a big SB insulates the protocol from Black Swans and the need for flops, which are devastating from a user trust and experience perspective. And the optic points are important, too. If outside observers see us “self-policing” with a big SB, it puts us in good stead vis-a-vis our centralized and decentralized competitors.

1 Like

Oh yeah, I 100% agree on this point. For some reason, I thought “serious reputational hit” was referring to the Collateral Auctions rather than the Flop Auctions. Apologies.

1 Like

No worries, I should have been clearer.

1 Like

Bias should be toward burning when mkr is undervalued, bias toward building the buffer if token is overvalued. Find a method to value the token based on annual/quarterly revenue, earnings, revenue/earnings growth rate, and surplus buffer (risk can be included in this formula). Base the burn ratio on estimated intrinsic value compared to market price. If the surplus buffer gets larger than necessary, it can be used to buy back mkr when it is extremely undervalued.

All governance should be aimed at increasing intrinsic value of the token.

There is no way we will agree on the parameters, on which such valuation would depend. You would get way too high valuation band.

1 Like