Signal Request: Limit Vaults to a maximum % of global DAI supply

Hi everyone,

I just created a new proposal to set a trigger for when DC adjustments can be voted on. With that in mind I also wanted to create another proposal that ensured that certain vaults didn’t get too big relative to other collateral types to ensure the safety of the protocol. For now I think we only need to worry specifically about USDC, WBTC, PAX, and USDT as they are all centralized.

It’s important to note before voting that if another collateral type like ETH or KNC was getting too big we have the power to pause DC increases through emergency action.

Should we limit WBTC-A Debt Ceiling to a specific max % of global DAI supply?

  • Yes
  • No
  • Abstain

0 voters

If Yes, what should that % be?

  • 15%
  • 20%
  • 25%
  • 30%
  • Abstain

0 voters

Should we limit USDC-A Debt Ceiling to a specific max % of global DAI supply?

  • Yes
  • No
  • Abstain

0 voters

If Yes, what should that % be?

  • 5%
  • 10%
  • 15%
  • 20%
  • Abstain

0 voters

Limit PAX Debt Ceiling to a specific max % of global DAI supply

Should we limit PAX Debt Ceiling to a specific max % of global DAI supply?

  • Yes
  • No
  • Abstain

0 voters

If Yes, what should that % be?

  • 2.5%
  • 5%
  • 10%
  • 15%
  • 20%
  • Abstain

0 voters

Should we limit USDT-A Debt Ceiling to a specific max % of global DAI supply?

  • Yes
  • No
  • Abstain

0 voters

If Yes, what should that % be?

  • 1%
  • 2.5%
  • 5%
  • 10%
  • 15%
  • 20%
  • Abstain

0 voters

Polling will end 2020-09-08T04:00:00Z

Next Steps

If any/all polls pass, I will submit a pull request to have these added to a governance poll the following Monday. Once ratified, governance facilitators will be responsible for ensuring automatic debt ceiling adjustments do not exceed the agreed upon percentage.

Edit: Had to reset polls to add abstain.

3 Likes

@ahbartsch Should add abstain for the % polls too.

Added them. Please revote if you already did.

Great to see another initiative and the pervious one for ETH. I though think that debt ceiling shouldn’t defined as % of global DAI supply. In our models when we perform risk assessment of collateral, debt ceiling is in majority defined by exchange volume of the collateral and stress testing downside price returns. So DAI supply clearly can not be a input from risk perspective and for some assets it would be just unwise to increase debt ceiling.

But I do agree we need a better or new framework for ongoing changes to debt ceiling. Ideally we would simply provide a curve for each collateral type with a DC cap and corresponding risk premiums on its utilization. This isn’t an easy task though and requires a lot of ongoing modelling for each collateral type, but we hopefully get there soon or find some kind of shortcut.

2 Likes

Thanks for the insight. From a risk perspective could you recommend tweaked to my methodology that would make it more feasible? Setting a max dc threshold based on exchange liquidity?

I think the answer lies in our ability to have more regular assessment of applicable DC plus SF, but unfortunately exchange liquidity is only one of the many inputs to measure this. Also, now when most of DAI minted is being farmed by same users, I see much less risks in terms of liquidation outcome as many vaults could be unwinded quicker before liquidations happen. Which means exchange liquidity actually plays just a bit smaller role, but is still very important.

Since we are understaffed to run these models for each vault type constantly and make assessments for new collaterals we should strive to calculate DC and RP numbers in a more practical way. Give us some time to come up with more practical solutions.

Polls end tomorrow!

1 Like

Based on the results of this signal request we will only be asking to automate DC raises for ETH-A and USDC-A. Because of this, the only parameter that has to be included as a PR for on-chain voting is to add the stipulation:

If USDC-A DC > 20% Global DAI Supply, no DC trigger will occur.

Although the other results aren’t actionable at this time, it’s great to have data on what the community thinks of the other collateral types and will be helpful feedback for future debt ceiling discussions and even MIP17 work. Thank you to all who participated!

1 Like

Having discussed this with Aaron, I think that these polls are a little too lacking in participation to take as a strong signal at this time.

I’d like to see the proposal refined some and gather a stronger consensus before moving forward on-chain.

1 Like

To repost my message from the other thread. I think due to their complex nature and the current state of the ecosystem, it is best to revisit this at a later date. I hope we can instead focus on getting the DC IAM up and running which would make this action unnecessary.

1 Like

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.