The Surplus buffer

hi everyone,

Anyone know what happened to the surplus buffer, it was nearing the 30 mil but it drop now below 20mil usd.

There are a lot of liquidations currently, and the bidding is currently often 0 DAI. So the bad debt in the Surplus Buffer. Might be corrected if the bid are higher. It’s more liquidation than Black Thursday.

We might lose 10M DAI or have a 1.3M DAI of profit (and burn some MKR today!). Next hours are critical.

The surplus buffer this morning:
image

The surplus buffer a few minutes ago:

image

You can track the situation here => makerburn.com

Anyone can bid here =>

2 Likes

hi,

Thanks for the update. i be looking at the situation .

What’s causing this liquidations? I don’t see big price crash.

hi,

I think the buffer protected the price, If we had to resort to the mkr token to cover the loses we seen a price impact.

It’s mainly one vault, but a big one always flirting on the verge of disaster. The rest of the vault is now safe. You only need to be below the liquidation price once and people are liquidating you.

Someone will have a bad morning.

5 Likes

It seems like a large ETH-B vault getting liquidated from:

ETH-B usually at 50M max debt ceiling but it’s now only at 37M debt

Ouch. Someone lost a couple of million today.
Makerburn is currently registering minus 135 days until the Surplus Buffer is full.
Will be very glad to see Liq2.0 activated for ETH.

actually adding ETH to Liq2 is already planned - there is a poll on it running this week

actually so far the liquidations of this vault are running quite promising. we might end up with some nice profits and some flaps

1 Like

the system working and doing its thing. Good test . After it completed we can see how it went.
Interesting day for the system and bad day for the holder of the CDP.

image

We are now making a small profit from it. Might be improved as we get to the close of the auction. It will be time to bid for flap auctions.

Today we might burn significantly.

9 Likes

Yes problem with Eth b vaults it’s different risk parameters. Have to adjust as auction end time gets closer. You see now system surplus at 800k and rising

Let me try and understand the logic employed by @makerburn and also write it down here for the benefit of others.

  1. Vault is below liquidation price and hence a part of their collateral is put up for auction.
  2. Since no bids are in yet, we assume the collateral will be sold for 0 DAI and hence the surplus buffer has to bear the cost of this “loss” but it isn’t an actual loss yet.
  3. As bids come in, the highest bid on collateral exceeds the DAI borrowed using that at some point and that’s when the surplus buffer is no longer going to be at a loss. As bids go even higher, the buffer goes into profit.
  4. When the auction concludes, we finally know the actualized profit (or loss if step 3 didn’t happen).
  5. If MKR is burned, the buffer goes back down.

Is that about right?

On a side note, 50k DAI went for 22.16 ETH in a few auctions - that’s quite a lot lower than current ETH price! Guess we really need Liquidations 2.0 to remove this inefficiency.

Yes, you are right. As long as there are to tend-bids the deficit is covered by the System Surplus. With every step in the tend-auction, the amount of DAI covered by the auction rises - reducing the deficit in the System Surplus.

If an auction get tended to DAI 44247 the loss for the protocol is covered, if it moves above that we also collected the penalty fee - resulting in a bonus for the System Surplus.

In dent phase the DAI is then auctioned for the collateral.

So to sum it up: in tend the auctions finds out how much of the DAI will be covered by the liquidators instead of using the System Surplus, in dent we find out how much of the colateral will still be in the vault after the liquidation.

All of the liquidations we saw yesterday

  • had a tend phase fully covering the deficit
  • most of them even collected the full penalty fee (went to dent phase)

As keepers have to commit to a price for some time the bids in dent resulted in more collateral taken away from the liquidated vault. This is one of the main advantages of Liq2: Instead of bidding on a price that gets settled later, they basically buy some collateral and can recycle that in a single TX.

Still noteworthy, that even in Liq2 a liquidation starts with covering the deficit through the System Surplus - as long as the outstanding DAI is not covered by a auction

1 Like

While makerburn doesn’t implicitly consider liquidation auctions at all, this is essentially what happens. The system surplus value comes from vow-contract (0xa950524441892a31ebddf91d3ceefa04bf454466) by comparing the amount of dai dai() to the amount of unbacked debt sin(), so it’s not specifically makerburn making these considerations, but MakerDAO itself.

Btw, I’m sorta overlooking the current liquidation system, since it’s being replaced. But for anyone who haven’t noticed, makerburn already supports Liquidations 2.0 with LINK currently and YFI getting activated shortly!

12 Likes

makerburn is simple and clear. Would it be possible to give some suggestions.
Like adding a daily burn bar chart. And fees collected in a 24h time span and the mkr remaining in supply.
So new people can understand better how mkr and dai work.

1 Like

SB has dropped a couple of million but I don’t see any big liquidations kicked. Did I miss something?

It’s Protocol Engineering Core Unit payday.

5 Likes

Question, With the new proposal accept , where when the SB reach 30 mil. The SB will be cap at 60 mil and there be a 25% fee used to burn mkr. How will it show here Dai Stats. Like will you guys make a seperate mkr burn and the SB so we can see how much goes to which account.

The SB will increase progressively to 60M and we expect that the burn will be around 25%. There will be burn phase, then SB increase phase in which we don’t burn, burning again …

1 Like