[Signal Request] Increase ETH-A DC IAM `gap`

LFW: By @Aaron_Bartsch and @ultraschuppi 's powers combined, here’s an ETH-A DC-IAM gap signal request.


Hi Everyone,

With the current DC-IAM configuration on ETH-A we are having gap set to 30MM. Looking at the chart of DAI-from-ETH-A compared to the DC, we can clearly see that we hit the DC a couple of times.

image

Additionally, the currently running onchain poll for raising ETH-A line looks like we are going to increase it a lot. I guess we need to adjust the gap to keep up with the demand.

Another way of solving this issue is lowering the ttl but given the current gasprices, this is not a good option.

Pros:

  • risk of hitting the DC before ttl is done is getting lower
  • we are not artificially slowing down appetite for DAI-from-ETH-A
  • decreasing the potential for griefing attacks
  • more headroom - if a true bullrun starts, we can satisfy a surge on demand a lot better

Cons:

  • we are not artificially slowing down appetite for DAI-from-ETH-A
  • higher exposure to OSM timing attacks, but since OSM risks are much lower on ETH-A - probably not a big risk

Should we increase the `gap` of the ETH-A DC IAM?

Please choose all options you agree with.

  • No Change (30M gap)
  • +10M DAI (40M gap)
  • +20M DAI (50M gap)
  • +30M DAI (60M gap)
  • +40M DAI (70M gap)
  • +50M DAI (80M gap)
  • +60M DAI (90M gap)
  • +70M DAI (100M gap)
  • Abstain/Show Results

0 voters

Next Steps

If a result other than “No Change” gains a majority I will create an on-chain poll to vote on March 1st, 2021. This signal request will run until 2021-02-25T05:00:00Z.

5 Likes

Looking from the consensus sentiment of risk on this boards, I would say the IAM parameters are working as intended, limiting the growth of ETH exposure while letting borrowers know when new Dai will be available, avoiding the embarassment of ETH DC being capped out for several days.

That being said, my personal opinion is that ETH is our favorite collateral, which took us through the bear market and to 1 billion Dai, and it is silly to be picky with it.

2 Likes

@Aaron_Bartsch Personally I think +30M is too low.
The increases in this one are more juicy

1 Like

The 30m ‘gap’ we proposed few weeks ago was based on historical daily debt growth observed for ETH-A. It seems debt growth has only picked up since then and 30m gets hit more frequently than expected. But this is partially also because Maker rates are currently much lower compared to current competitive and market rates in my opinion (both Compound & Aave borrow rates are over 10% for ETH).

2 Likes

I’ve merged @ultraschuppi 's signal with this one as it was a little better for information and links. Thanks to both @Aaron_Bartsch and @ultraschuppi for creating an ETH-A gap signal.

3 Likes

Based on @ultraschuppi 's signal request there seems to be a lot of appetite to raise the gap even more than the proposed choices. Therefore I will be redoing the poll and increasing the end time. Sorry for the inconvenience but I’d rather we not have to do this twice too soon.

4 Likes

I am happy to hear about downsides of having a lot of headroom on the gap. IMHO we should be prepared for a serious surge on demand on DAI-from-ETH and currently only ETH-A is capable of delivering that. But for that, we do need a much higher gap.

Starting with 100M seems not unreasonable, but on the long run we might need a lot more.

Remember that having a 100M gap gives us the possibility of 200M more DAI a day. We’re not there yet so we can revisit that maybe later in the year if things start to get really crazy.

We don’t need to set a gap as close as possible to the current demand, instead we should have a gap that allows us to feed the demand if a real bullrun starts.

That’s fair. I’d personally like to meet somewhere in the middle.

Personally, I think there’s a bit too many options for a signal request.

Yea I probably could have gone every +20M instead :sweat_smile: Oh well!

1 Like

Nuff said

3 Likes

Just wanted to bump this poll for anyone who hasn’t participated. It will be open for 3 more days. Thanks!

2 Likes

I will now be closing this poll. It looks like “+50 (80M gap)” has won with a majority of votes and so I will create a pull request for this to be added into next week’s voting cycle. Thank you to everyone who participated!

1 Like

if you deduct “Abstain” 70MM would have 50% too…

I’ll defer to an executive decision by @LongForWisdom but I think going with the result with more votes will be fine in this case. No need to IRV a 10M difference especially when @Primoz suggested a 100M gap in ETH-C. Would the correct wording be plurality of votes?

+50 (80M gap ) would be the larger majority. Plurality is defined as something like:

“the number of votes cast for a candidate who receives more than any other but does not receive an absolute majority.”

I also don’t think it’s super worthwhile to run IRV for this. I’m aware this is inconsistent with what we’ve done before, but IRV adds a level of complexity to polls that I don’t really like unless there is a clear value add. Will Maker Holders appreciate having the option between a 70M and 80M gap, or would it be something they just shrug over and rank them arbitrarily? I’m not sure, but I suspect the latter.

If people really want (and think it is worthwhile) to have IRV for this then reply and let me know.

4 Likes

On-chain polling for this SR has now passed and will be included in tomorrow’s executive. Thanks again!

3 Likes

The executive that includes this parameter change is now up! Please consider voting for this and other changes! https://vote.makerdao.com/executive/parameter-updates-and-onboarding-rwa-001?network=mainnet#proposal-detail

4 Likes