[Signal Request] Offboard unprofitable collateral types

Overview

I want to follow up on the topic raised by @NikKunkel in the G&R-Call on 2021-05-20 about unprofitable ilks.

Since MCD we onboarded a pretty decent amount of collateral types, and it shouldn’t be a surprise that not all of them got the adoption we hoped for. We already have a couple of collateral types where the gas we pay to keep the oracles running exceed the SF we collect.

There are a couple of steps that we could do before actually stopping the oracles for the ilks we want to offboard. I really believe we should start doing something ASAP so with this Signal Request I want to focus on the easiest step:

  • check if the community is fine with offboarding specific ilks

All other steps like raising the SFs, fiddling with the LR (which is debatable) to force liquidations or at some point just shutting down the oracles to not further update the oracle prices are not part of the Signal Request - but I would be happy to hear your ideas and thoughts about the next steps.

Pros:

  • less gas-costs for updating the oracles
  • less complexity for oracles team
  • less complexity for PPGs

Cons:

  • fewer ilks ¯\_(ツ)_/¯

Shall we offboard USDT-A?

image

  • No
  • Yes
  • Abstain

0 voters

Shall we offboard UNIV2-AAVEETH-A?

image

  • No
  • Yes
  • Abstain

0 voters

Shall we offboard KNC-A

  • No
  • Yes
  • Abstain

0 voters

Shall we offboard UNIV2-LINKETH-A

image

  • No
  • Yes
  • Abstain

0 voters

Shall we offboard COMP-A

image

  • No
  • Yes
  • Abstain

0 voters

Shall we offboard BAL-A

image

  • No
  • Yes
  • Abstain

0 voters

Shall we offboard LRC-A

image

  • No
  • Yes
  • Abstain

0 voters

Shall we offboard ZRX-A

image

  • No
  • Yes
  • Abstain

0 voters

Shall we offboard BAT-A

image

  • No
  • Yes
  • Abstain

0 voters

Shall we offboard MANA-A

image

  • No
  • Yes
  • Abstain

0 voters

Shall we offboard AAVE-A

image

  • No
  • Yes
  • Abstain

0 voters

Next Steps

The Polls will run until 2021-06-17T13:00:00Z; their outcomes will either result in on-chain-polls assuming the outcome of the polls deems it necessary or its intermediate results are going to be taken to another initiative - another signal, onchain-poll or urgency executive.

6 Likes

Did you mean TUSD or USDT? The label doesn’t match the graphic, and one has an outstanding debt of ~26.5m and the other of ~700.

Edit: Thanks for fixing

3 Likes

I believe there will be clear one and not that clear one.
Can we have a some sort of market analysis on the non clear one?

As said on the other thread, there aren’t many others asset to source more supply anymore. May be it is worse to try something before getting rid off them completely.

Do you think that USDT in the Maker system will start to be used more with the possibility of PSM, because a PSM would be more effective than directly opening Vaults?

What’s your thought on talking with relevant collateral types to see whether they want to put more into their own collateral to better utilize and also bring more revenue? If their option is having their collateral removed from Maker and unlikely to be on onboarded in near future, it might sense for some of them to use it more to avoid that.

3 Likes

Maybe worth exploring. I’d feel more comfortable with them just paying our Oracle costs than being huge vaults themselves— since they can mint the tokens themselves to dilute us if they suffer a governance attack or just get greedy

Minting doesn’t change anything unless it is massively within an hour time frame.

I mean, it is not desirable but that is for the whole market, the token price will go down and they will get liquidated.
What it matter is liquidity, capital market and don’t be exposed over something like 10% of the capital market and based on the liquidity.

I mean that applies to other collateral types as well. They might get greedy or attacked and abuse the system.

But liquidation also brings revenue as long as they are liquid, no?

1 Like

Paying Oracle cost is an interesting point. Do you know what’s usual annual cost for each?

1 Like

Depends on gas costs. I think Nik has suggested a 200k-300k/year ballpark figure unless the Oracle for that asset is already being used (like for an LP pair)

1 Like

Is that an accurate accountantcy, I think you deserve better :slight_smile: .

I believe this year was a special year. Do they have addresses from where the transactions are sent?

Also, it is not because gas wasn’t cheap and the oracle not optimized that should be the norm.

Thanks for sharing. I guess when gas price skyrockets, it indeed can be quite expensive

Hopefully Optimism and/or Arbitrum allows vaults a lower break-even point for customers and Maker itself.

1 Like

But going back to it, I definitely think especially for more DeFi tokens, might make sense to suggest them to use their treasury token as collateral to mint Dai similar to what YFI team is doing. Increase in revenue, more diverse collateral, more efficient use of treasury, everyone is happy (in theory)

1 Like

This was a full round of ‘Abstain’ for me.
Before offboarding collateral types, after all the hard work of first onboarding them, I would very much prefer to see a cost breakdown with the gas cost included. Prior to that we do not really have any data to go by. Also we could maybe refer edge-cases to Risk and see if the collateral type could have a lower debt ceiling in exchange for more beneficial Vault parameters.

8 Likes

Is gas cost the primary reason for them being uneconomical due to oracle updates?

As I pointed out in the past Maker should not even think about onboarding collateral that can’t generate more than 50K/yr in fees when something like 2% of the Market Cap is deposited (getting 10% of any coin is unlikely imo). This means coins that have at least 50M caps and ideally we want stuff like 500M and above.

We want to start at the top and work our way down was ignored and here we are - basing getting rid of things based on oracle costs and completely ignoring the other costs associated with bringing these on. Really want to urge Core groups to focus on the low hanging fruit vs. just going after everything as manpower is limited.

While I won’t touch USDT with 10 ft pole personally I don’t think MKR ever gave USDT a fair chance at a 8% SF. Up the LR and reduce the SF and see if we can capture any part of that huge market.

I pretty much said YES dump everything except ZRX and USDT for above reason. I don’t know why I think ZRX should stick around vs. AAVE or BAL. So if I were to be consistent I’d say dump ZRX and see if anything can be done with USDT to get some activity. I have felt since the beginning USDT with the high fees would never get a single deposit.