Raising the ETH debt ceiling

Can we raise it to at least 300M @cyrus? We saw that some whales mint $20M+ from one vault.


Thanks for the fast exec vote for 260M, @cyrus but can we have another raise now? Why do we have to deal with the DC every week? It’s not like someone will dump 100M DAI and drop the price (it would be nice if someone did that).

Please put a poll or raise the limit further to 400M+.

I feel like I may have explained this to you before Bit, but on the off-chance I’m misremembering here goes.

Leaving a large gap between the current debt and the debt ceiling presents a risk if the price of ETH drops significantly in a short amount of time due to the delay on the oracle price updates. Specifically, (due to the 150% ratio,) the price would need to drop by more than 33% between oracle price updates.

In this case, users can exit to Dai using the higher OSM price and abandon their collateral, which the system then has to liquidate at a loss. The upper bound for this risk is the difference between the current debt and the debt ceiling. Because of that, it’s beneficial to keep them as tight as possible given market conditions.

I know you’ve done signal requests previously, that would be the avenue to make something like this happen outside of the Risk team recommending the change.

The risk team recommending action is the fallback if unexpected market changes crop up faster than governance is able to react through the non-emergency processes, that said, the emergency actions aren’t meant to be a replacement for governance managing the ETH-A ceiling themselves.

If governance want to see a larger buffer maintained, then that would be something to signal for.


I also agree that it would make sense to have a proactive conversation about what kind of free gap to leave available in the ETH DC.

I think those moments where the DC is hit during an ETH rally is the kind of situation where a lot of new users might be lost since there isn’t enough DC to service them, which is a shame and might add up in terms of lost adoption if we keep doing it every week.


Maybe there are ways to adjust our parameters to reduce the risk of this happening in the first place? Like updating oracles every 15 mins instead of every hour? We will likely have to make this adjustment later anyway, since lowering the CR on eth and other tokens(something we’ve talked about in the forums a little bit) would exacerbate this issue.