[Signal Request] Increase Uniswap DAI-ETH, USDC-ETH, & DAI-USDC Debt Ceilings + add DC-IAM

Overview

The Oracle and Smart Contracts Domain Teams have reviewed the preliminary ABDK audit results for the Uniswap V2 LP Oracle contract. No critical or high priority issues were found that would have adverse effects on any of the LP tokens that the Maker Protocol currently supports. The Smart Contract / Oracle Domain Teams are making the appropriate fixes in the Oracle contract and will resubmit the result to ABDK to compile a final version of the audit report that will be published. Subsequently, the current instances of the Uniswap V2 LP Oracles will be replaced with the updated version.

In the meantime, now that we’ve confirmed the Oracle contracts are safe, the community can consider raising the debt ceilings for one or more of the LP tokens. It should be noted that Aave launched support for LP tokens today and will likely eat away at our marketshare without us raising the debt ceilings of LP tokens.

The primary counterpoint to raising debt ceilings is that there is liquidation risk for LP tokens. There is no significant primary liquidity for LP tokens. While we do see some participation in liquidation auctions for LP tokens, there is no guarantee that all keepers have updated their setups to support LP tokens.

After discussions with the Mandated Actors including the Risk Domain Team we feel that the liquidation risk is worth the upside of competing for LP token marketshare and are recommending the following changes.

Changes:

Increase UNIV2USDCETH-A debt ceiling to 20% of pool liquidity => 50M
Increase UNIV2DAIETH-A debt ceiling to 20% of pool liquidity => 30M
Increase UNIV2DAIUSDC-A debt ceiling to $50M
Add DC-IAM to UNIV2USDCETH-A (final params pending, 10% of DC w/ 12 hour gap)
Add DC-IAM to UNIV2DAIETH-A (final params pending, 10% of DC w/ 12 hour gap)
Add DC-IAM to UNIV2DAIUSDC-A (final params pending, 10% of DC w/ 12 hour gap)

It should be noted that the DAI-USDC pair has none of the aforementioned liquidation risk due to the DAI and USDC Oracles being fixed to $1.

Reasoning

USDC-ETH and DAI-ETH were selected because of their high liquidity, significant LP fees, as well as proven demand within the Maker Protocol evidence by their debt ceilings being fully utilized. This bodes well for their potential to generate significant amounts of Dai.

DAI-USDC is more of an experiment. It has low LP fees (5.3%) which will likely decrease as liquidity is added to the pool. However, there is potential that once a critical threshold of liquidity is reached that new trades begin to route through the DAI-USDC pair, increasing volume, and ultimately LP yield. It is relatively safe to jack up the debt ceiling to 50M relative to the current $24M in the liquidity pool due to liquidations being turned off.

Please vote on ALL options you would support in an onchain-poll.

Adjust the UNIV2USDCETH-A debt ceiling to:
  • 50M
  • 40M
  • 25M
  • 10M
  • No Increase
  • Abstain

0 voters

Adjust the UNIV2DAIETH-A debt ceiling to:
  • 30M
  • 20M
  • 10M
  • No Increase
  • Abstain

0 voters

Adjust the UNIV2DAIUSDC-A debt ceiling to:
  • 50M
  • 40M
  • 25M
  • 10M
  • No Increase
  • Abstain

0 voters

Deploy a Debt Ceiling Instant Access Module (DC-IAM) for UNIV2USDCETH-A?
  • Yes
  • No
  • Abstain

0 voters

Deploy a Debt Ceiling Instant Access Module (DC-IAM) for UNIV2DAIETH-A?
  • Yes
  • No
  • Abstain

0 voters

Deploy a Debt Ceiling Instant Access Module (DC-IAM) for UNIV2DAIUSDC-A?
  • Yes
  • No
  • Abstain

0 voters

Next Steps

The Polls will run until 2021-03-22T10:00:00Z ; the outcomes will result in a on-chain-poll assuming the outcome of the polls deems it necessary.

13 Likes

Thanks @NikKunkel , fantastic news!
Congratulations!

I have one question: I have not understood if you think it would be a good idea to activate the IAM? Generally speaking the IAM is a good idea, but in this specific case (low liquidity, dangerous liquidation procedure) I am not sure.

For the moment I voted “Abstain”. But if you express a positive opinion, I will change.

thanks!

2 Likes

Ending on Monday doesn’t really leave us time to setup polls for that same week. If the results are positive and this has had sufficient attention by Friday 19th, I’ll close it then and we’ll prepare the on-chain polls for Monday.

Alternatively we could push it to the week after, but I suspect people would rather see this sooner rather than later.

Edit: Have pinned this until tomorrow so it gets enough eyes.

5 Likes

Going forward how will we look to resolve the lack of interest by Keepers when liquidating these UNI LP vaults? Any thought? Ideas.

Assuming volatility hits 40%+ and the risk of IL increases/comes-into-play, would Keepers have an attraction to this asset class?

Friends,

This is fantastic. I notice DCs are being set as a function of the size of total pool liquidity.

Is the cap on the total DC for these types over the long term always a function of total liquidity? If so what %?

One of the reasons I ask is Maker obviously wants to grow the Dai issued from these Ilks as quickly as possible. However I am wondering to what extent that will be constrained by total pool liquidity.

If there is a fixed % will that change if Liq2.0 proves effective?

Its not hard for me to imagine a universe where the majority of Unilp tokens for some pairs are held in MakerVaults…

1 Like

Hayden (Uniswap founder) has hinted that Uniswap V3 should be released soon. If the V3 release is similar to the V2 release, then a majority of assets should migrate from V2 to V3 over the course of 1-3 months after the release of V3. This timeline can be accelerated by additional liquidity mining. The V2 liquidity mining program included many of the pairs we are discussing here.

Has the risk team thought through the case where this happens and V2 LP tokens become substantially less liquid (potentially 90% drop, depending on protocol and liquidity mining incentives)? Is there any concern?

8 Likes

I don’t know of a secondary trading market for V2 LP, wouldn’t liquidations just lead to redeeming the underlying assets? Which shouldn’t be impacted by liquidity

6 Likes

I believe you are correct. After thinking through it, I do not believe there is any negative outcome for the Maker system during a Uniswap V2 to V3 migration.

2 Likes

Confirming proposed parameters by @NikKunkel. As mentioned on G&R call today, the liquidity of underlying tokens (ETH and USDC) for these pairs does justify DC increase to levels proposed. We also feel much more comfortable regarding auction risks after hearing that auctions UI that have shown no bids for some of UNI LP auctions had a bug. Plus we may have potentially mitigated this situation with higher ‘beg’ value and we still haven’t seen a larger liquidation event that would likely attract more keepers to bid.

Regarding IAM-DC, here are suggested values:

UNIV2USDCETH-A
ttl = 12h
gap = 5m

UNIV2DAIETH-A
ttl = 12h
gap = 5m

UNIV2DAIUSDC-A
ttl = 12h
gap = 10m

5 Likes

Only thing is that I don’t think that DC increases need to be paired as a liquidity percentage. ETH-DAI is actually less risky than ETH-USDC and easier for Keepers to liquidate.

Anyways, we can always increase line limits in the future.

4 Likes

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.