Expedited Executive Vote: 2020-09-14

Edit: Vote is now live

Hello everyone. As most of you are aware, an emergency signal request was posted over the weekend and has resolved in favour of action.

Given this outcome an executive vote will go on-chain on Monday. However, for a number of reasons that are detailed below, the parameters included in the on-chain executive will not directly match those presented in the emergency signal poll.

Emergency Signal Results

Winning Emergency Signal Option

  • 32/55 Voters
  • ~58% Majority

USDC-A LR: 101%
USDC-A DC: 200 million
PAX LR: 101%
PAX DC: 100 million

Losing Emergency Signal Option

  • 25/55 Voters
  • ~45% Minority

USDC-A LR: 105%
USDC-A DC: 200 million
PAX LR: 105%
PAX DC: 100 million

What will be included?

Executive on Monday

USDC-A LR: 110% -> 103%
USDC-A DC: 100 million -> 200 million
PAX LR: 120% -> 103%
PAX DC: 5 million -> 30 million
Box Parameter: 30 million -> 15 million

On-chain Poll on Monday

Subject: LR Ratio on USDC + PAX combined.
Vote Type: Ranked (IRV)
Options: 101%-105%
Caveat: If the Monday executive passes, an outcome of 104% or 105% will not continue to executive.

Why have we made changes?

  • Governance might not be properly informed about the scale the system might be affected at 101% LR for USDC-A and PAX. - @Primoz (more in a reply shortly)
  • The previously planned USDC-A LR on-chain signal poll lets us drop to 101% on Friday 18th if MKR Holders vote in favour of it on-chain. This poll will be expanded to include PAX.
  • It is difficult to go back up to 102% or 103% after dropping to 101%, so it makes sense to have this as an intermediate step.
  • The emergency poll was heavily split between 105% and 101%, as was the preceding signal request. 103% represents a compromise between those two options.
  • PAX was just introduced last week and we only started to test demand. As we showed in our collateral evaluation, PAX still has a very low presence in DeFi where most of DAI price arbitrage will probably be performed. This is why we are limiting proposed increase to only 30m, instead of 100m. - @Primoz
  • Box parameter change to 15m DAI was initially intended to go into Executive on Friday, but based on new information from keepers about the level of their capital reserves, we would want to perform that change as soon as possible. - @Primoz
13 Likes

As someone who was torn between the two options this seems like a reasonable compromise with polling for reduction during the week. I support these changes.

3 Likes

I would have voted for 103% if it was one of the options

As noted above, my feeling was that governance might not be properly informed about the scale the system might be affected at 101% LR for USDC-A or PAX. It is expected to have heavily increased stablecoin exposure at LR 101%, probably between 150-200m DAI debt minted on USDC and PAX from day one. Furthermore, arbitrage trades could be performed by single a Vault user by using iterative flash loans and collecting huge profits by only spending gas fees. We would then hope that this user repays its loan when DAI < 1.01 (net of fees). If not, Vaults would need to be manually liquidated, hopefully during liquidation 2.0, so Maker doesn’t incur too much slippage and carries potential loss. Note that this is not the main reason for why it was decided to make an executive with 103% LR, but I just wanted to stress this, before a lower LR is proposed to an Executive vote (based on a Monday’s poll).

Also note that my personal preference is to decrease LR slowly and be more aggressive on collateral onboarding with assets such as cUSDC or yyCRV, where we basically achieve similar stablecoin exposure (more other protocol risk though) but Maker could charge much higher SF. We know that USDC-A won’t be collecting much fees at low LR.

11 Likes

We know that it is a good idea to enlist assets such as cUSDC or yyCRV as collateral, but how long will it take? One month is too long, can two weeks?
The current PEG issue is increasingly hindering our development. I found that some users in the market are giving up DAI and choosing USDC instead. If the collateral we choose is beneficial to PEG, we should complete the work as soon as possible.

2 Likes

Awesome. Yes, it is difficult to go back from 101%. 103% CR is still a big move forward for stabilizing the peg, and we still have options on the table to do more action on Friday.

There is absolutely zero chance we can get cUSDC or yyCRV done in a month. Furthermore, on the governance and risk call last week the community seemed to indicate it wanted the domain teams to stop focusing on onboarding regular collateral types but focus on more long-term items like real world assets at the expense of not doing any collateral onboarding for 2-3 months. The community can’t have their cake and eat it too. I understand it’s frustrating. At some point it has to be acknowledged that domain teams have a limited amount of resources and have to strike a balance between short term and long term priorities.

18 Likes

Yeah, that was my understanding also.

When RWA was voted the priority, this also meant bigger short term reliance (and risk) on USDC (and other stablecoins). I am also not thrilled, but there seem to be no other choice. Well, maybe the only other choice is ‘bad’ peg.

2 Likes

Sell 140M DAI will lower the price to 1.01 and not all the generated DAI will used to short.

1 Like

Your screenshot shows 14m, not 140m :slight_smile:

1 Like

thank you for the explanation. In other words, the PEG tool that can be activated quickly is only LR = 101% (USDC, PAX, TUSD and other stable currencies), while other solutions require a long waiting time.

1 Like

A few detail questions:

  1. What time will the vote go live?
  2. How long will be it be live?
  3. If it goes through, when will be the on-chain implementation?
  4. What are the requirements for it to pass? (How many MKRs needed?)

I think 101% or 102% LR should go into the vote. Looking at the current condition where 1 DAI = 1.04 (4%) USDC. Making the ratio to 101% will easily pull DAI below 1% or 0.5%.

Increase in ETH price is the optimal solution for DAI stability and at that time the USDC stability fee could be raised really high which will push everyone to close their position.

Worst case scenario if 101% is not reliable for Maker system longer term the governance can setup stability fee really high >100% (making sure USDC-A never be usable again) and release USDC-C after the DAI is in a stable state.

2 Likes

They are aimed at 16:00 UTC, but this can often slip by a few hours.

It’s a continuous approval vote, it’s live until it passes, or until people decide it isn’t going to pass and want to pass something else.

After it passes, there is a 12 hour delay.

it needs to beat the current proposal which has 107k MKR. Note that much of this MKR will probably swap to the new proposal. So you need ~55k to swap to pass it.

Executives can’t use ranked voting. We had to pick one thing. As described there will also be a ranked on-chain poll with the options 101% - 105%, the results of which could be implemented on Friday.

So 103% on Monday, possibly 102% or 101% on Friday, depending on how the vote goes.

11 Likes

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