Maximally Expand Credit Lines

TLDR
Expand all current credit lines to 2M each. Give new credit lines to LUSD, LINK, and FRAX.

Maximally Expand Credit Lines
This proposal seeks to expand the credit lines for silo markets currently receiving credit lines to the maximum amount supported by XAI trading liquidity. Further, 3 more assets currently not receiving credit lines will have credit lines extended to them while meeting restrictions from XAI and base asset trading liquidity.

Current credit lines are listed below.
500K XAI in cbETH, wstETH, DAI, and USDT
400K XAI in gOHM,
100K XAI in OHM

This proposal expands all the above credit lines to 2M XAI each.

Further, these 3 assets do not yet have credit lines and this proposal seeks to extend credit lines to those:
2M XAI in LUSD
1M XAI in LINK and FRAX

1 Like

Hey Puppy!

Glad to see your proposal here- I completely agree with expanding lines. Here are some recommendations on my side based on the liquidity as I see it now so we don’t have to run any more credit extensions on these assets after this, would love feedback.

We should raise liquidity from-to the following:
gOHM - 500K → 1.5M
cbETH - 500K → 1M
wstETH - 500K → 1M

Extend New Lines:
LUSD - 2M
LINK - 500K (Liquidity can sometimes dry up so recommend we are a bit conservative)
FRAX - 1M

1 Like

Hello!
Great that you agree that credit lines should be extended.
My proposal was not for extending credit lines somewhat, it was to expand them maximally. According to your own analysis, all the assets mentioned can take a full 2M XAI each as they have sufficient base asset liquidity. LINK is the exception which can only take 1.5M XAI. I propose we expand the credit lines to those values.

The benefits:

  • Devs need to raise credit lines less often if done in bigger increments saving valuable dev time
  • More XAI will lead to interest rates, which will drive growth of TVL and usage
  • As the credit lines get borrowed up, the DAO will see higher medium/longterm profits

Cons:
-Lower short term profits for the DAO

  • We can extend credit lines with ease. It takes minimal efforts from the dev team. I don’t a reason why we should increase credit lines before CL’s utilization exceeds 50%.
  • OHM/gOHM share the same liquidity depth and therefore credit lines for both token assets should be treated as one. There is a borrow demand for OHM rather than collateral, given the rebase APR the asset generates when it is staked. I recommend we focus on expanding gOHM’s credit line.
  • I don’t agree to extending 2M XAI fully to volatile assets. In a market crash scenario, we can assume the following:
    • ETH price drops by half.
    • Liquidity dries up by 50%-60%, increasing price impact drastically.
    • Cascading liquidations lead to increased price impact.

I recommend:

  • Increase gOHM credit line to 1.4M XAI
  • Increase wstETH credit line to 1M XAI
  • Increase cbETH credit line to 750K XAI
  • Extend 1M credit line to FRAX
  • Extend 500KM credit line to LINK
1 Like
  • “We can extend credit lines with ease.”
    I am getting different information regarding how easily it is for devs to extend credit lines. In the proposal discussion discord, it was stated by a team member that it is something that would take away precious dev time.

  • “I don’t a reason why we should increase credit lines before CL’s utilization exceeds 50%”
    I have explained the reasoning, and given the pros and cons of expanding credit lines, and those apply even when credit line utilization is below 50%. You can disagree with me about my thesis (credit lines would lead to lower interest rates, which would drive borrowing), but I have proposed that we put it to a vote so the DAO can decide.

  • “I recommend we focus on expanding gOHM’s credit line.”
    Agreed

  • “I don’t agree to extending 2M XAI fully to volatile assets”
    You describe that silo would take on systemic risk from giving the full 2M XAI to volatile asset silos. If that is a risk, then I agree that we should be more cautious when extending to volatile asset silos. I do, however, think the current methodology for credit line worthiness seems insufficient and improvised/unprofessional. For example, is the maximum credit line extension for a silo (with good quality oracle and sufficient base asset liquidity) 2M? After this conversation, the 2M XAI amount seems to only apply for stable assets, is that right? If so, what safety margin does a volatile asset require? (What happens if a stable asset depegs, which happens occasionally? should stable assets be treated as volatile if there is depeg risk? How do you quantify that risk)
    I’m not getting any clear numbers, and it does not give me confidence this process is being handled well.

  • Did you edit your answer? I believe you had a point regarding cbETH but have changed your opinion in the last few hours.

  • You missed out LUSD completely

  • " * Increase gOHM credit line to 1.5M XAI"
    As gOHM is a volatile asset and you say it should be given 1.5M, then I’m assuming that’s an appropriate level for volatile asset silos. (You edited your answer to 1.4M from 1.5M)

Therefore I recommend:
1.5M XAI for the volatile asset silos [gOHM, wstETH, cbETH, FRAX, LINK]
2M XAI for stable assets silos [LUSD]

Ps. stop editing your answers. It’s ruining the discussion, and it’s not how one discusses in good faith.

  • Effort is irrelevant - if it needs to be done, core team will do it. I explain below why we should go steady slow.
  • Your thesis is likely correct but it is irrelevant to Silo. We are not a stablecoin project that makes money on minting volume. Our DAO implements a clear strategy as far as XAI which aims to generate revenue to cover costs associated with maintaining deep liquidity at the peg plus a margin. The DAO needs to make 2-3% on each credit line. Lowering interest rates is always easier than increasing them. You are rushing into conclusions prematurely without accounting for the time factor.
  • The process is professional and transparent. We have a hard cap on how much CL each qualified asset can get. The Cap is calculated based on simulating the dump/pump scenarios. We need to maintain a balanced, deep poo for XAI Curve pool. When the pool grows deeper, we can perhaps increase the 2M cap. A balanced, deep pol means: XAI is available to buy back at $1, or very close to $, to liquidate large collaterals - mainly collaterals of volatile assets. The risk assessment model is untested yet and that is why we ought to increase credit lines gradually, monitoring for unknown unknowns. Yes, CL numbers we have started with are arbitrary, but that is acceptable as long as we can adjust them going forward.
  • Stablecoins are safe to extend CLs to. LUSD is perhaps the best stablecoin out there given its backing and decentralization. Reputable stablecoins deviate from the peg but never depeg. LTVs protect against minor deviations. Stablecoin are not volatile. There is not any known risk extending large CLs to each reputable stablecoins - it is more of wanting to target 2%-3% APR, hence the DAI/USDT experiment.
  • gOHM/OHM credit lines are actually one since one is derived from the other. Combined, they can get 2M. We are only choosing to cap at 1.5M because of the aforementioned reasons. to be careful. Together they get 500K now.
  • Please be kind in the way you communicate. I don’t engage in hostile conversations. I deleted and posted within 10 mins, I did’t think users read my comment.

I’ll start by responding to this as it’s an important point. There are multiple ways to turn a civil conversation into a hostile one. One can use a rude tone/language, one can be closed off to outside opinions and not acknowledge the merit of those opinions, and one can use bad faith tactics.
You have not been open to the merit of my idea. We have been discussing this proposal for about a week, and it is only now that you give any concession and even that you do begrudgingly and dismissively “Your thesis is likely correct but it is irrelevant to Silo.” We are disagreeing on some points, but calling my arguments irrelevant is rude and closed-minded.
You posted a reply, then either deleted it and posted a new one, or deleted it and then edited your reply. Either way, it seemed like a bad faith tactic (Which reply am I responding to? Why did you have to change your reply/opinion within a short time?). If you say it was an error on your part, I’ll give you the benefit of the doubt and accept that.
Now, I very much want us to have a civil conversation. But for that to occur we both have to show humility, and that includes being open to that one isn’t always right and others’ ideas might be worth pursuing. Hostility breeds more hostility. Be nice, and you’ll be treated nicely.

“Effort is irrelevant”
I think we should strive to do things efficiently. That includes doing larger credit line expansions less often as compared to many small credit line expansions more often.

“We are not a stablecoin project that makes money on minting volume”
I always said that one of the cons of my proposal is that there will be “lower short-term profits for the DAO” (see my posts above). Now, please try to understand what my point is: It is okay for the DAO to make less profit in the short-term so that we can grow our TVL/protocol more quickly. You’re argument seems to be that this is something that we shouldn’t do as any credit line expansion should be funded by profits the DAO makes from interest rates. That’s very fiscally responsible, but I do not think fiscal responsibility should always be our highest priority. Our TVL is now below 30M USD. It is extremely small. We should focus on growth, and fiscal responsibility is something that can take second place for a short time while we grow the protocol.

" Lowering interest rates is always easier than increasing them" I think you are mistaken, or I’m missing something. Expanding credit lines will drop utilization and therefore interest rates (driving demand). It is not hard to later increase interest rates. Actually, the DAO needs to do nothing at all if utilization picks up, interest rates would go up by themselves. If the CL remains unused, then a proposal to remove/decrease the CL can be easily implemented, as the CL remains unused. So either it requires no effort, or small effort.

“You are rushing into conclusions prematurely without accounting for the time factor.”
Please remain humble, I am not rushing to a conclusion, so don’t assume so. I have given this a lot of thought. I have thought about the time factor, and as I said, decreasing interest rates only affects DAO profitability in the short term, but increases profitability in the long term.

“The process is professional and transparent.”
Fantastic, sounds like the current process gives an answer of 2M XAI per silo. This proposal keeps within that limit “1.5M XAI for the volatile asset silos [gOHM, wstETH, cbETH, FRAX, LINK], 2M XAI for stable assets silos [LUSD]”

“monitoring for unknown unknowns” And who decides when enough monitoring has been done, You or the DAO? If it’s the DAO, let’s put this proposal to a vote.

Aiham, as a general point, we are having a difference of opinion. You want the DAO to take it slow and steady because it limits risk. I think we should be taking on more risk with the benefit being faster growth. Both approaches have merit, and there is no right or wrong answer because nobody can look into the future and say which approach leads to the better outcome. And it doesn’t matter because it’s the DAO that decides what strategy is followed and it should be voted on by the DAO. So, when are we gonna have a vote on it?

I am not dismissing your opinions, they are important, I happen to disagree with them. You are welcome to spin a proposal to affect your ideas - I say that with open heart and mint. DAO will decide.

A few points:

  • There is so much work. I cannot attend to every conversation in a timely manner. Sometimes I post half-baked opinions (and delete or edit), and I am realizing that is less effective - thank you for pointing that out.
  • DAO making revenue means we have more incentives to spend on attracting TVL, and encouraging borrowing in the protocol (borrowing XAI and other assets). That might create a flywheel effect. I am afraid we we lower IRs, we don’t make revenues to run the buyback/incentive program. Yes, we will have more XAI borrowed, but I have serious doubts volumes will go up to levels where we make the same level of revenues that we are making now. You say we increase IRs if that would be the case, I say it is better to keep what we have for enough time to see the impact of current strategy and adjust based on results. Strategy is already in motion.
  • This is what I meant with time factor: 1- Reach: Our reach is limited. We might think the model is not working because it is flawed whereas it might be because community doesn’t know about it. We are working on discovering the reason - and it might be both. 2- XAI is new, new code, new project, etc. Time = trust in DeFi, I am for exercising patience and discipline than shifting gears.
  • This is not my DAO - you or anyone should participate and affect changes. Like anyone in the DAO, I will always express opinions loud and clear against initiatives that I don’t agree with. I will always do that publicly. This is too my right.
1 Like