9,9 is the strategy of staking OHM, lending it in return for DAI, use DAI to buy and stake additional OHM. This can create increased demand for OHM and higher returns for the daredevils. It also can accelerate losses and cause cascading sell-offs in the OHM token.
9,9 is essentially about increasing the effective yield per asset while accepting a higher risk level. For Olympus as a protocol though, the old 9,9 has proven to be a double-edged sword.
What if we could enable 9,9 while keeping the protocol shielded from any downside?
We could do this by turning the cards on what asset is acting as collateral for the loan. In old 9,9 OHM was the collateral and exposed to liquidation risk and sell pressures. In the new 9,9 OHM could be sitting on the other side of the pair and the liquidation risk/sell pressure would instead be shifted to the second asset.
This new 9,9 concept extends what is described in the post “Collateralized Convertible Bonds” posted previously. Please read that first. In that post, one drawback mentioned was the lack of double yield on the collateral. As a working name lets call the solution to the problem “Uncollateralized Convertible Bonds” (UCB).
For a bond buyer of an UCB, the experience is identical to buying the CCB. The bond buyer always pays with OHM/gOHM and it goes into locked staking. UCB price decays same as other V2 bonds.
The bond issuer however, is going to issue the bond without any collateral. Well, not quite true – we need collateral in some sense. So, how?
Option A: Intangible collateral
A prototype thought was to build a reputation based (credit score) system, enabling whitelisted actors with brand value to issue non-collateralized convertible bonds. In order to preserve their brand value/reputation/credit score they would need to deliver the promised assets ahead of maturity or their credit score would be implicated through some metric such as ohmSeconds and probably penalty fees (0.33% per time unit). But this would all be built on trust and brand value. The better your credit score, the higher your premium will be. But… Time is probably not right for this setup yet.
Option B: Pre-existing collateral on the blockchain
Another option would be to seek pre-existing, lockable collateral elsewhere. Where do we find such collateral? It exists in the form of notes for those who have bought a not-yet-expired v2 bond with term shorter than the UCB that is being issued; AND it exists in the pools established to support CCB’s. Both locked notes and pooled assets carry the potential to back uncollaterized convertibles, thus enabling the reversed 9,9 strategy. To keep it clean, lets disregard the notes and focus on the pool as our choice of collateral.
Use case for bond issuer (BI):
1. BI has previously deposited 100 ETH as one sided liquidity into a CCB pool.
2. BI owns another 40 ETH, and expects to collect 10 ETH worth of yield using these 40 ETH outside of Olympus ecosystem.
3. BI checks the reserve requirement for ETH on Olympus UCBs; it is 50%. Since BI holds 100 ETH in CCB, he is able to issue a UCB worth up to 50 ETH.
4. Once BI issues the 50 ETH UCB, his 100 ETH becomes unwithdrawable from the CCB pool.
5. Upon UCB maturity, redeeming ETH holds the higher value versus the gOHM but the BI has not supplied the promised collateral, which constitutes a Failure to Deliver and the position becomes open for liquidation.
6. A liquidator spots the opportunity – provides the 50 ETH collateral to the UCB and the claim on 100 ETH in the CCB pool is transferred from the issuer to the liquidator.
Alternative 5: Upon UCB maturity, redeeming gOHM is the high value option for the bond buyer. No necessity for the issuer to provide the collateral.
Summary
As this use case illustrates, we were able to issue a UCB without providing anything but pre-existing collateral. In this case, we have kicked up the leverage 50% allowing us double yield on a pre-existing asset. Furthermore, this has allowed a lockup of 50% extra OHMs. If a liquidation occurs – all OHM continue to stay locked for term durations. This means there is no sell pressure on OHM in case of liquidations. We manage to 9,9 without the negative repercussions for the protocol.
Note:
In the use case, we used ETH to ETH as collateral cuz its simple. But in real life application, the collateral/exposure could hopefully be recalculated to OHM on a portfolio basis and the reserve ratio continuously monitored. The liquidation opportunity illustrated at maturity above, could open up at any time during the term if the reserve ratio is not be upheld due to price movements. Also, as mentioned, notes are also a potential source for pre-existing collateral thanks to the existing push/pullNote functions of the v2 contract.