Turn ETH liquidations off until the peg is restored

Here’s another “radical” solution to the peg problem but I firmly believe that there is no easy solution that can bring the DAI price down.

This measure is, of course, not always applicable but considering that ETH is far from overvalued by any calculation and it is relatively stable / growing in the last 6 months - now it might be a good time to use it.

Maybe there are alternatives to when to turn the ETH liquidations on - i.e. in X months or until ETH price is $XXX or…


Can you provide (any of) such calculations ? :astonished:

I don’t think there are any widely accepted valuation models for eth.

It’s definitely out of the box thinking which is great. However who are we to determine the fair value of an asset? That should be left to the market to decide, why solely ETH? We would be entering dangerous territory. If it falls below liquidation do we still have to pay at fair value internal calculations when vault owners want to cash out? Seems way to risky, sends a dangerous precedent and it’s not our job to guarantee market valuations.

I’d have to search but I don’t think we should trust any calculations anyway. It’s kind of common sense. If the value of ETH network (MCap) stays at $27B (compared to easily accessible world money at $37T) then ETH and DAI will have no impact on world economy. Such a complex system as ETH is not sustainable if it’s just a playground for nerds without considerable business use.

Why don’t we all go and buy ETH then? Because ETH might have fatal flaws (technical/human/governance/adoption/legal…) but in that case DAI will fail also.

Also, if ETH loses its #2 spot by market cap permanently to another smart contract platform - ETH and DAI are dead because no critical mass adoption will happen.

may be we should start by lowering the LR from 150% to 130 - 135%

a lot of people (myself included) choose to use Compound or Aave because I am more comfortable having more space from getting liquidated


I agree that this is a much more sensible and conservative change that accomplishes about the same thing.


That’s a step in the right direction but it will result in very little improvement. We will have to repeat the step and then eventually stop at 101%.

As I said, it will not be enough (just like raising the DC for USDC and wBTC). The demand is too strong.

We don’t know whether it will be enough or not, but unless you have facts and figures, I don’t think you can say for sure that we will have to keep repeating this process. We are not going to want to lower the LR too much though because we do not want to take on that risk. We are still facing the consequences of black Thursday.

We’ve been talking about dropping LR to ~135% anyway, as this would make us more competitive in the defi market. It would be nice if auctions were updated before we did that though, since we would be more able to guarantee overcollateralization of dai even in extreme market events.


I have mentioned before that anytime the Dai price is over the Peg, liquidation of assets to buy Dai from the market is counter productive. Liquidations in that case serve to increase Dai demand further. It would make more sense to sell Dai until the Peg is corrected and then to liquidate if it was necessary at that point. OR for MakerDao to “buy” the collateral in the vault for freshly minted Dai it self as it is cheaper to do so than to buy from the market at inflated Dai prices.

Had this been the case during March there would have never been $0.00 bid auctions. It is ridiculous that the system gave collateral away for free like that.

1 Like

Yes but in a flash crash Maker gets stuck in a depreciating asset.

Edit: Unless we apply the 13% penalty on the purchase and sell it. It should cover any loss

That is better than selling for potentially $0.00 or paying 15+% (Dai at $1.15+) above the market rate in most cases. So long as the collateral being purchased is diversified I don’t see it as a problem. Everything gets easier as there are more diversified and liquid assets available for the system to use. Right now that isn’t the case yet (which is why this period is hard to navigate).

If Dai was over peg and ETH falling for example: Why add pressure by buying more Dai from the already demand overloaded market? Instead I say MakerDao should sell Dai for other assets and if the price of Dai gets back in line then go ahead and buy it for the ETH.

If not, trade the ETH for other assets as well if exposure to ETH was deemed too high and continue selling Dai until the price is back to Peg before unloading ETH for Dai directly. MakerDao profits by selling Dai into a market over the Peg price and loses in the opposite scenario… Why lose when MakerDao can gain?

Yes but you are making recurrent a non recurrent event. Before printing dai out of thin air I’d prefer to use the surplus to buy the assets at a discount, dump them on the market for mkr and burn it. This should do since keepers knowing maker can participate should not bid at 0.

That is the idea. MakerDao “buys” the assets for Dai from the Vault paying back the debt and collecting the liquidation penalty. Afterwords if collateral is higher than levels that MakerDao is comfortable with it would auction the debt for other assets other than Dai to rebalance the collateral in the system such that it is still at proper levels. It would have a large profit to work with in order to accomplish such aims through the liquidation penalty.

Those “reserves” that get built up during liquidations occurring while Dai is over the peg price in the market would be emptied if and when Dai ever comes back to being under the peg price.