There's some ideas that I've been musing on a bit lately and felt worth airing just to see what folk think. This is likely to be rather long, since there's a fair bit of context involved.
The core of it revolves around the question: Is there sufficient utility / favourable risk/reward in tighter integration with certain partners, or in some cases, in-house implemenation of a wider range of defi primitives / building blocks?
Stated differently; what if Olympus could become a decentralised financial giant, offering a wide range of services to end-users that have synergistic effects, creating additional positive feedback loops for the protocol as a whole. Perhaps, if such things have merit, they could in time become a core part of the Olympus suite, alongside the core pillars of staking, bonding & protocol-owned-liquidity.
Of course, one could well argue that this would be far too much to take on under one roof, and that we'd be better served to "do one thing, and do it well", but all the same I think this would be a very interesting topic for discussion, even if it may not lead anywhere. I am also but a layperson, so there are quite likely angles and risks on this that I've failed to consider.
At least to my observation, there's a strong desire amongst different cohorts of Ohmies for leverage (for the more risk-tolerant) and for downside protection (for larger holders who might be pretty comfortable with what they have, and would like decentralised, ideally tax-efficient ways to lock in some portion of their position while remaining staked, not selling (i.e. (3, 3), not (-1, -1)).
There are certainly existing options for some of this at present; at least on the leverage side of things, with Rari Fuse, Abracadabra and in the future Fiat DAO (perhaps also Aave/Compound/MakerDAO down the track). For downside protection however; things are more complicated.
There's work in progress with getting existing defi insurance protocols on-board to protect specifically against risks stemming from contract bugs or economic design issues, but to my knowledge, there's not anything that Ohmies can use right now for this. In future, perhaps wsOHM/gOHM could be on various decentralised options platforms, and likewise, perhaps we may one day see wsOHM/gOHM on something like BarnBridge's smart alpha; but all these third party solutions eventually run into issues of reaching critical mass / sufficient liquidity to achieve their potential.
With that context out of the way, I come to the core of what I've been thinking about: What if some such services were implemented under the Olympus banner itself, with tight integration with the DAO and treasury, albeit in ways that minimise risk to the protocol.
Concepts:
#1 An integrated money market similar in function to compound/aave/rari, with Olympus itself in a privileged position of first-in-line liquidator.
To be very clear here, this would **not** involve lending out treasury assets; the backing of OHM would not be compromised. Instead, as with existing solutions in this space, borrowing and lending would be peer to peer between Ohmies. The key benefit of integrating something like this, rather than leaving it up to third-parties, would be to give Olympus priority (but not obligation) to liquidate defaulting positions.
Various forms of collateral could be deposited, to be borrowed by others including sOHM (or wsOHM/gOHM - but in either case, OHM or its wrapped variants would **not** be borrowable through this facility) but otherwise limited to assets that can be bonded.
Rates would be determined as a function of utilisation, again, similar to existing products, with the majority of the interest accrued by borrowers being captured by lenders, however a *small* portion of this interest would be captured by Olympus itself, bolstering treasury growth beyond what is already captured through bonding. As with similar offerings, there would be a liquidation penalty (for the borrower) / incentive (for the liquidator) of some percentage.
The consequence of this, is that by granting the protocol itself the ability to essentially front-run liquidations, desirable collateral (recall that collateral would be limited to bondable assets + OHM/wsOHM) could be acquired by the protocol at a discount (i.e. a net gain for the treasury).
If a position collateralised by OHM/wsOHM/sOHM is liquidated by the protocol, it'd be equivalent to a protocol-initiated buyback (at a discount for the protocol), and naturally any sOHM/wsOHM acquired by the protocol would immediately be unstaked and held as OHM alongside existing protocol-owned OHM.
If the protocol itself is uninterested in liquidating a specific position (e.g. in normal circumstances, it would likely not be desirable for the protocol to liquidate OHM-collateralised positions), then it can be left up to external liquidators who are incentivised to do so, due to the aforementioned liquidation incentive.
#2 (Mostly) Liquidation free, self-repaying loans via isolated wsOHM-OHM collateralised bonding.
Alchemix's core product consists of borrowing pairs where one deposits an asset and may draw a debt in corresponding synthetic token that is pegged (via 200% collateralisation) to the underlying. As a result of this, liquidations due price movement cannot occur as debt and collateral assets have identical price action. In such a setup, liquidations can occur only when the yield on collateral falls below the interest accruing on debt for an extended period of time.
There has been prior discussion about sOHM on Alchemix, but as discussed in the linked thread, there are a number of (difficult to resolve) issues with handling such as facility externally. In particular: It can be challenging to get sufficient liquidity between the synthetic and the underlying to allow leverage (due to the opportunity cost of doing anything with OHM other than staking it), and in general, even if a synthetic were not used, none would be willing to loan out OHM for anything less than the staking APY, negating the utility of the whole concept.
Consider however:
With these points in mind, Olympus could offer the ability to borrow OHM, overcollateralised (e.g. 200%) by sOHM (or wsOHM/gOHM). A modest interest rate would be charged on loaned out OHM, far below the staking APY, but non-negligible all the same (I imagine this interest rate would be a function of available idle OHM in the protocol and the current rebase rate). Yield on deposited collateral would go to reducing debt, only increasing collateral once debt is neutralised. The implications of this include:
- sOHM locked in such a position to service a debt is removed from circulation, since it cannot be redeemed while the debt remains, reducing the circulating supply (since, with 200% collateralisation, at most 50% could be borrowed; the net impact on circulating supply would be negative)
- The DAO would capture a somewhat greater portion of new issuance, coming from the interest rate charged on OHM debts; in essence this would be redirecting a (very small) amount of staking issuance to the DAO.
- In the event of a liquidation, due to the collateral's yield falling below the debt's interest rate for an extended period of time, the debt to the DAO would be negated by seized collateral, and remaining collateral could be redeemed by the depositor.
This enables a number of use-cases:
- Ohmies wishing to leverage up can do so in a way that does not risk chain liquidations as we've seen with existing third-party offerings.
- Ohmies wishing to de-risk can partially hedge their position (in a way that may be more tax efficient, as it does not require them to sell sOHM they've held long-term), while continuing to earn staking APY by borrowing OHM against their sOHM and immediately selling it, reducing their exposure to OHM's price movements by up to 50% (i.e. it does not allow them to be net short).
#3 Downside protection / liquidation-free leverage via senior / junior tranches.
This is less a concept for duplicating in-house, and more of a partnership opportunity; perhaps one that could be particularly tightly integrated, with its functionality surfaced in our own frontend.
BarnBridge currently offer a product (Smart Alpha) that implements precisely this; pools are created for a given asset, priced by some reference asset (for example, wETH-USD). Users can deposit the underlying (e.g. wETH) into either tranche and receive a derivative asset that for the senior tranche, has a dampened upside, but a downside protection to a defined level, or for the junior tranche, leveraged exposure.
When the price of the underlying declines, the junior tranche experiences an amplified downwards price movement until the senior's level of downside protection is fully covered. As the level of downside protection for the seniors and level of leverage for the juniors is determined on a per-epoch basis, as a function of the relative liquidity of each tranche, liquidations do not occur in the traditional sense; even in the event of the underlying going to zero, the junior tranche's maximum loss is 100% - no debt exists in this setup.
It's also worth noting that the derivatives minted from depositing into either tranche are themselves ERC-20 tokens, so they could be used elsewhere (i.e. in LPs, as collateral in money markets, etc) while still enjoying the reduced downside (senior) / leverage (junior).
Unfortunately, from the perspective of Ohmies, there are two issues with using BarnBridge's offering:
- wsOHM is not, at present, a supported asset on their platform.
- Even for assets that are supported, the utility of a given pool is heavily dependent on the available liquidity for its comprising tranches.
Considering this, at the very least there's an opportunity for a partnership here; both for wsOHM to be made available on the platform (with a pool priced relative to Dai, and perhaps also a pool priced relative to ETH), and to improve the liquidity of said pools.
Olympus could utilise a portion of DAO-owned OHM to provide liquidity to both tranches, leaving Olympus with equivalent amounts of both bb-senwsOHM and bb-junwsOHM derivatives (i.e. equivalent exposure to simply holding what was deposited), which could be simply held as is, or paired with OHM in bb-senwsOHM/wsOHM and bb-junwsOHM/wsOHM LPs.
As entry/exit to/from Smart Alpha pools is not liquid (entry and exit can only occur once weekly, at the transition of epochs), there would likely be decent demand for liquidity on third party LPs, which Olympus could own and collect fees upon. There could also be particular benefit in protocol-owned-liquidity on something like Balancer where a single pool can contain more than two assets; e.g. a Balancer pool of OHM/bb-senwsOHM/bb-junwsOHM, but this in of itself would be a new thing for Olympus, as we currently have only Uniswap and Sushi LP/SLPs for protocol-owned-liquidity.
Tying into the first concept put forth in the post, bb-senwsOHM and bb-junwsOHM could be accepted as collateral in the hypothetical integrated money market, with appropriate collateral weights (greater for then wsOHM itself for bb-senwsOHM due to the dampened downside, lesser than wsOHM itself for bb-junwsOHM due to the increased volatility resulting from leverage).
