Rebalancing Mechanism Enhancements

In light of recent events, I’d like to aggregate several community ideas from several different threads onto one thread to discuss next steps.

Ideas were mostly from @arch @Ken_M @AwaxJago and @Uglyrage. Perhaps I missed some, didn’t do a deep dive into the history of the forum, just operating off memory.

Overall, to enhance the current rebalancing mechanism, there are a few key features that were discussed:

1.) Rate limiting over time periods - only a certain amount of bnUSD should be able to be minted/burned through rebalancing over a certain time frame.

2.) Asymmetric Peg - Reverse Rebalancing does not need to be the same as regular rebalancing. In fact, it should likely be set at such a high level that it would be impossible to push somebody into liquidation. Based on liquidation LTV, this can be implied with some simple math.

3.) Organizing Borrowers into a list sorted by LTV, then only rebalancing the highest LTV accounts - This is from Liquity.org, our closest comp from ETH ecosystem. I’m on the fence as to whether or not this is better than evenly spreading across all accounts. They don’t really have this option since ETH is so expensive. Of course, this would only apply to regular rebalancing, not reverse. Liquity, fwiw, has no reverse rebalancing at all. Their LUSD and climb to any height above $1.

These ideas all came from the community, myself included. If anybody has any other thoughts, please share on the thread. The goal is to make the borrower experience on Balanced as good as possible without sacrificing too much of bnUSD stability.

2 Likes

This all sounds like steps in the right direction.

#2 “Asymetric peg” especially would have saved alot of people.

1 Like

I would like to expand on point 2. I think it would be ok to have no upper limit to the peg since a rising bnUSD price should be incentive enough to mint and sell it.

1 Like

I agree. And ideally with more adoption more users and more collateral types bnusd becomes more and more stable even during harsh times.

In Balanced, secured loans place a greater burden on borrowers than other DeFi.

We believe that it should be managed separately so that borrowers can choose to

rebalance, so that borrowers who allow rebalancing can receive more rewards, and

borrowers who do not allow rebalancing can only manage their own loan risk.

1 Like

If I am reading this correctly, rate limiting over time periods will addresses two problem areas. First, insulating positions from changing drastically in a short amount of time. And second, the problem of having a “hard edge” to the peg.

Does anyone have insight into how to achieve this?

I’ve thought about a similar idea for achieving this goal for the stability fund with a cool down mechanism. But there are probably better ways.

I will do some homework on this.

1 Like

My current leading idea is to use a percentage of bnUSD supply as the threshold. This gives borrower’s the most accurate idea of how their position could change on a daily basis. For example, if 5% of bnUSD supply were chosen, that would mean your debt couldn’t be rebalanced by more than 5% per day.

Another thought I had was a percentage of the sICX/bnUSD liquidity pool (or other pools when multiple collateral types happen), which essentially limits the amount of additional liquidity that could come from borrowers at a given price point, however, this doesn’t provide a borrower with as much predictability.

1 Like

I guess figuring out that percentage will take some thought. Would the percentage be manageable on a vote (similar to threshold, maybe even combine them)? As I imagine there might be adjustments needed if the peg drifts too far for too long.

2 other variables to consider:

Deeper bnUSD / sICX pool wil make rebalancing less frequest. But when it happens, more is needed for the same result.

Partial liquidations while not per se helping with rebalancing, would help alot to prevent a rebalancing <-> liquidation loop

1 Like

Asymmetric Peg would be nice.

I think it has been discussed before but i really think a variable peg would also be very interesting. Something like every time a re balancing event happens move the peg further and the slowly move it back towards the original when the reblancing slows down.

Another thing i am curios about is moving rewards from bnusd/sicx pool to loans. There might be that more bnusd is minted along with more bnusd is added to the pool since there are more bnusd available. Thus there might be more optimal reward distribution there to increase depth of both pools.

Partial liquidations and a bidding system for the liquidation is a MUST.

There’s talk that an upper limit is not needed.

In my mind, a high enough level that wouldn’t push loan’s into liquidation would operate similar to no upper limit. But I don’t know if that is necessarily the case.

Would having an upper limit of 15% enough to limit reverse rebalancing from pushing people into liquidation?

In this scenario, when rebalancing happens, loans and collateral would increase, but collateral would increase with an extra 15%.
100 bnusd added to loan; $115 value in ICX added to collateral.

I want to share this post from @arch (from another thread) that has views that are pertinent to how we move forward. Specifically with 1 & 2.

Yeah that’s definitely an interesting point about how the value of the debt is fixed to the oracle price, not the true price of bnUSD

So does this mean that borrowers who have the highest loan to collateral ratio would be rebalanced first?

Would it be in cohorts, or would 1 account get rebalanced until it is no longer the highest, then build a cohort as the top gets brought down?

Here is how it could work:

Jack, Janet, and Chrissy are in the below situations

Jack

  • Debt = 1,000 bnUSD

  • LTV = 30%

Janet

  • Debt = 2,000 bnUSD

  • LTV = 40%

Chrissy

  • Debt = 1,000 bnUSD

  • LTV = 50%

A 2,000 bnUSD rebalancing comes in.

  1. Chrissy’s entire position is closed, all 1,000 bnUSD of her debt would be repaid.
  2. Then Janet would have 1,000 bnUSD closed off.
  3. Jack would have nothing happen to him at all.

This could be spread out a bit more to say that only x amount of your position could be effected. Or at least take the top x borrowers with highest LTV instead of just the top 1.

Keep in mind this does not effect reverse rebalancing.

My reasoning for posting it was to highlight the danger of both rate limiting over time and an asymmetrical peg combining to keep bnUSD outside of the peg for an extended period of time.

If top end rebalancing is set at 10%, and rebalancing begins at $1.10, but the rate limit is reached before the value is back within the peg, and it climbs further away each day.

I don’t know what values to use, or if this is a big concern. I thought I’d bring it up to as something to consider.

I just want to go through some examples, to make sure I am illustrating the core concept correctly, and give some rate possibilities for feedback on.
Please correct me if needed.

Edits: I have focused on reverse/positive rebalancing. I have added volume of new bnUSD minted for rebalancing, based on 10,000,000 bnUSD supply.

Format:

Heading line:
5% (rate limit) 500,000 bnUSD (added to loans):
First line:
1000 (value of collateral in bnUSD), 300 (value of loan in bnUSD), 30% (LTV)
Second line is when rate limit reached
1015, 315, 31%

Rate possibilities:
5% 500K bnUSD:
1000, 300, 30%
1015, 315, 31.03%

1000, 350, 35%
1017.5, 367.5, 36.12%

10% 1 M bnUSD:
1000, 300, 30%
1030, 330, 32.04%

1000, 350, 35%
1035, 385, 37.19%

15% 1.5 M bnUSD:
1000, 300, 30%
1045, 345, 33.01%

1000, 350, 35%
1052.5, 402.5, 38.24%

20% 2 M bnUSD:
1000, 300, 30%
1060, 360, 33.96%

1000, 350, 35%
1070, 420, 39.25%

25% 2.5 M bnUSD:
1000, 300, 30%
1065, 375, 35.21%

1000, 350, 35%
1087.5, 437.5, 40.06%

33% 3.3 M bnUSD:
1000, 300, 30%
1100, 400, 36.36%

1000, 350, 35%
1115.5, 465.5, 41.73%

40% 4 M bnUSD:
1000, 300, 30%
1120, 420, 37.50%

1000, 350, 35%
1140, 490, 42.98%

50% 5 M bnUSD:
1000, 300, 30%
1150, 450, 39.13%

1000, 350, 35%
1175, 525, 44.68%