Borrowing is what keeps Anchor sustainable in the long term. The growing value of aUST under Earn comes from the growing outstanding liabilities owed to Anchor by borrowers. More revenue is generated from borrowers’ provided collateral at minimum 200% value.
For some, high collateral requirement for low interest rate (or at least interest offset by high rewards) is great, as it allows for a lot of exposure to the collateral bAsset underlying, whilst providing the ability to borrow a little extra UST and do with as they please.
However, for other potential borrowers, it would be more convenient to require a relatively low amount of collateral (still over 100%), even if it costs more interest.
I believe allowing the latter type of borrowing will still be able to generate revenue for Anchor, but also reach an untapped market of borrowers, thereby improving Anchor’s sustainability.
There are many ways Anchor could go about this. I have many ideas, but I just want to get people thinking. If you’re interested, here are some of the things I’ve come up with: There could be multiple lending categories at different rates. Or a continuously recalculated interest rate based on a user’s collateral-to-loan ratio; a variable rate could help many borrowers’ liquidation fears (while still not eliminating liquidations altogether). There could be rewards based on amount of offered collateral rather than loan amount. There could be bAsset earnings paid to users who offer collateral but borrow under 25%. There could be Airdrop aggregation to those who offer their bAssets as collateral. etc.
I very much agree with this concept. Since it appears that borrowing demand will be the bigger bottleneck for Anchor growth, we should be looking into more opportunities to make borrowing easier and more attractive. The risk of liquidation is currently a large consideration on the decision to borrow and the amount.
We should only be liquidating borrow positions when they approach an LTV that poses a stability and solvency threat to the protocol. Is this LTV level really 50% for bLuna or is the 50% LTV requirement to ensure staking rewards on posted bLuna are large enough to service interest payments to UST depositors?
Perhaps there are 2 different LTVs for these two separate concepts (Solvency Risk LTV and Staking Rewards Shortfall LTV).
It’s destructive to the protocol to liquidate borrowing positions that aren’t actually threatening the solvency of the protocol. It’s possible that the Staking Rewards shortfall LTV could be considerably lower than the Solvency Risk LTV. If a borrow position only risks creating a cashflow shortfall, a penalty interest can be charged that would eliminate this issue.
For example:
The ‘Cashflow Shortfall LTV’ might be 40% but the ‘Solvency Risk LTV’ might be 70%.
Once a loan passes the ‘Cashflow Shortfall LTV’ (40%), the entire value of the loan would incur an incremental penalty interest rate of +10%. This would compensate UST depositors for the lower amount of staking rewards relative to the size of the loan.
Once a loan passes the ‘Solvency Risk LTV’ (70%), the loan poses a systemic risk to the protocol and would be put up for liquidation.
In light of the major crash in the past 24-hrs (which may continue), I see many users commenting on other platforms (e.g., telegram) that they’re discouraged from ever borrowing from Anchor, because it’s just too risky; being liquidated is just too big of a loss in these kinds of dips. And we seem to be in agreement that Anchor cannot sustain without Borrowers and Collateral. So, perhaps we need to more heavily consider better liquidation protection. I think it’s possible that a solution for this topic could also help address the liquidation issue. Something like an adjusting interest rate instead of a 50% liquidation threshold would at least help encourage more borrowing and thus will help Anchor keep its revenue.
People are also asking about reserve depletion, and encouraging collateral provision is another way to help sustain revenue.
I think that’s a great idea. The challenge will be to maintain some balance between the impact/threat of liquidation and discouraging users from maximizing their loan LTV and thus driving the loan distribution APR down.
But there has to be some middle ground where you can support the platform with a responsible loan and be able to sleep without worrying you’ll be liquidated overnight.
This seems like a good idea. Would it be possible to implement this? “Cash flow shortfall” and “Solvency Risk” will also provide greater transparency to the user about how their loan status is impacting the system and why liquidations of varying levels are necessary.
I agree with the premise that borrowers = assets; depositors = Liabilities. However, higher interest rates do not offset the deleterious effects of increasing the LTV threshold. For one, a higher LTV threshold increases the risk of ruin. Risk is not symmetrical. You can make a lot of money in the good times, but if you aren’t careful, you’ll go bankrupt in the bad. The collateral is too volatile to increase the risk of ruin during liquidation and de-leveraging events that are inherent to leveraged markets.
Plus, a higher interest rate will deter more borrowers and potentially lead to a decrease in interest-bearing assets, while interest-bearing liabilities are not changing - or worse, growing. We are witnessing this dynamic now with the utilization ratio dropping to 44% despite the net borrow APR increasing. Anchor’s value proposition to the core customer, borrowers, is simply not competitive with borrow APRs at 20-30% when Compound and Aave offer stablecoin loans for 4-7%. Plus, Compound and Aave have a distribution advantage on the (albeit inferior) Ethereum network, which has plenty of USD fiat gateways and wallet options.
Money Markets are a commodity business. Thus, for Anchor to be a massive platform, it needs to be a trustworthy low-cost leader. That means Anchor needs to offer loans at or below 4% profitably while paying 18-20% interest to depositors. What we learned in the last 48 hours is that to achieve these levels sustainably, Anchor needs a fortress balance sheet that can absorb losses while the collateral is collapsing in value. Let’s hope that loans grow before the yield reserve runs dry.
However, I have a number of issues with some of the statements you have made.
bLuna is too volatile to increase maximum LTV allowed.
This might be true. It could very well be the case that ‘Cashflow Shortfall LTV’ = ‘Solvency Risk LTV’. However, there has been no evidence presented by the dev team or anybody else of this hypothesis. It has been implied multiple times that the reason max LTV is set to 50% is to ensure staking cashflows are sufficient to service depositor interest accruals. I think it’s doubtful 50% is also the solvency risk LTV, since even on Ethereum lending protocols, where transactions are slow as mud, it’s very rare to see such a conservative liquidation ratio. Regardless, a study should be done to determine safe LTV (or if it’s been done already it should be published).
Higher interest rate would deter more borrowers
We are not talking about increasing the net borrow rate. We are just talking about decoupling the cashflow requirements of the protocol from the solvency requirements. There could be two ideal set of incentives to deal with these two distinct issues. For borrowers, an attractive rate is not useful if they aren’t able to get enough LTV from their assets. Today, it’s basically impossible to draw anything greater than 40% LTV because the risk of liquidation is too high and so the true ‘useful LTV’ is 35% and under. Providing a warning LTV range that charges a higher rate of interest allows time for borrowers to repay or provide more collateral without materially increasing protocol insolvency risk.
I agree that an evidenced-based exploration of 1. is warranted. I do not think the eth protocols are comparable as they have a diversified pool of collateral. The major benefit of more diversification is lower mark-to-market volatility, which declines proportionately to the square root of the number of assets. A basket of four assets would get you half the benefit of infinity assets. A protocol with single-asset collateral will support a lower max LTV, thus I would expect the Eth protocols to be more robust to volatility at this stage.
Thank you for elucidating your point regarding 2. A higher interest rate above a threshold LTV to prevent liquidations seems reasonable, assuming LTVs are within the safe parameters we discussed in point 1.
I like that there is discussion here.
I was thinking a formula something like:
Net APR = (bAssetBlockRewardRate * ProvidedCollateralAmount * 0.75) - (UtilizationRatio * LTV * BorrowedAmount)
w/ Liquidation @ 80% LTV.
*Before accounting for the early ANC Distribution to incentivize borrowing.
This way, for the borrower: some of the collateral rewards still go to the collateral provider; the interest rate increases as LTV increases to compensate for loans at high LTV; and there is still possibility of liquidation in cases of high risk loans, just much better than 50%. Anchor could start to see loans at a whole spectrum of LTV%s, and also much more collateral and borrowing.
To illustrate some scenarios, without accounting for the early ANC distribution to incentivize borrowing:
-A user who provides collateral but doesn’t borrow anything gets some of the rewards from their collateral–positive APR.
-A user borrowing at maybe 20% sees ~0% net APR, with collateral rewards and interest cancelling each other out.
-A user borrowing at 70% LTV has very high interest rate–negative APR.
Again, this is before adding the early ANC distribution, so for the next while, the Net Borrow APR would still be much more favourable to the borrower.
So users could borrow at 10% LTV, be earning rewards based on their collateral amount (instead of their borrow amount), and be earning revenue for Anchor all at the same time–it’s a win-win. Whereas, currently, a 10% LTV loan w/ Anchor is basically a loss for the borrower considering the opportunity cost of otherwise staking their assets.
I see this making users more likely to be willing to provide their bLuna as collateral, and be more likely to borrow.