• Proposal
  • OIP-26: Move OHM-FRAX pool to Uniswap v3

Summary

Migrate our Uniswap v2 OHM-FRAX pool to Uniswap v3 to make use of the added capital efficiency as well as the additional levers. When FRAX bridge over their liquidity to Arbitrum, we would migrate up to 50% of our OHM-FRAX pool to Arbitrum as well.

Proposal

  • Move OHM-FRAX pool to Uniswap v3

  • Leveraging v3’s concentrated liquidity positions, we would provide liquidity in a range from RFV per OHM (approximately 27 today, but we will use 25) to $1500; for reference, standard xyk (e.g., v2) pools implicitly range from 0 to infinity

  • This pool migration would be executed in several steps to ensure a smooth transition

  • The RFV of our OHM-FRAX LP position will go from $891,851 in v2 to $127,030 in v3. Overall this is a 1.9% decrease of our total Treasury RFV.

  • When FRAX move their liquidity to Arbitrum, we would deploy 50% of the OHM-FRAX pool to Arbitrum

Important Notes

  • We would end OHM-FRAX bonds due to the complexity for users and inability of our Treasury to take in v3 LP tokens

Motivation

One of the main selling points of Olympus are our bonds and locked liquidity. Our Sushiswap and Uniswap pools have been growing at an unprecedented pace, with the OHM-DAI pool having $82,849,405 of liquidity, and the OHM-FRAX pool having $18,426,25. Furthermore, both pools are entirely (99.5%) owned by us. This allows for predictability from the protocol side, as well as confidence from the investor’s perspective.

Given the essential role liquidity plays in the OHM ecosystem, it is imperative for us to  continuously improve its implementation and utilization. We believe Uniswap v3’s concentrated liquidity positions represent such an improvement. Regular liquidity pools are inefficient, due to spreading the liquidity equally over the whole range. Below is a picture of a OHM-ETH pool with just $1.39m in liquidity, earning more fees than we do right now on our OHM-FRAX pool which has more than 13x the liquidity.

With concentrated liquidity, we can squeeze out more capital efficiency and more fees out of what we already have. This benefits everyone, by having lower slippage on trades, as well as earning more revenue through fees.

TL;DR - Uniswap v3 offers advanced control over your liquidity, and for a protocol amassing large amounts of it, such as ourselves, this is extremely valuable.

Proposal

  • Move OHM-FRAX pool to Uniswap v3
  • Leveraging v3’s concentrated liquidity positions, we would provide liquidity in a range from RFV per OHM (approximately 27 today, but we will use 25) to $1500; for reference, standard xyk (e.g., v2) pools implicitly range from 0 to infinity
  • This pool migration would be executed in several steps to ensure a smooth transition
  • The RFV of our OHM-FRAX LP position will go from $891,851 in v2 to $127,030 in v3. Overall this is a 1.9% decrease of our total Treasury RFV.
  • When FRAX move their liquidity to Arbitrum, we would deploy 50% of the OHM-FRAX pool to Arbitrum
  • We would end OHM-FRAX bonds due to the complexity for users and inability of our Treasury to take in v3 LP tokens

Since we already have a pool on Uniswap v2, it would make sense to just move it to Uniswap v3. FRAX have already moved their main FRAX-USDC pool to v3 as well.

Now, because we have actual backing in the form of RFV,  we can use that value (27 RFV per circulating OHM, but we will use $25) as the bottom of the range on v3. The top of the range would be initially set at $1500, at our previous ATH. Both of these can be easily adjusted, and due to the technical know-how required and its time-sensitive nature, the Policy team would manage this liquidity position. We would do so with the following objectives in mind:

  • maximizing capital efficiency (utility of the pool + fee generation)

  • bottom end of range should always approximate the dollar value of backing

  • top end of range should be adjusted to avoid creating artificial price ceilings (initially set at ATH)

Arbitrum

Since Arbitrum doesn’t have much liquidity right now, especially in the form of stables, once FRAX moves their liquidity to Arbitrum, we propose that we do the same with up to 50% of the OHM-FRAX pool for starters. This should allow us to capture a chunk of that volume (and the corresponding fees) through our OHM-FRAX and OHM-ETH (proposed in OIP-25) pools. Along with that, it would of course be a significant boost to our Arbitrum liquidity so improving user experience there.

New RFV Calculation

Our Uniswap v2 pools have an RFV that is calculated based on the v2 constant product formula 2*sqrt(ConstantProduct). This formula assumes that our liquidity is spread over a price ranging from 0 to infinity. Of course, in v3 that is no longer the case, so a new formula must be derived to calculate the RFV of a v3 position. You can find a more detailed explanation in our Notion:
https://olympusdao.notion.site/Uniswap-v3-RFV-Deep-Dive-24ecf98997a04baf9743f4924255ab6e

We will provide a specific example of the current FRAX LP position below to show the effect on RFV and capital efficiency of moving from v2 to v3:

V2 Position: Current FRAX LP position in the v2 pool has 9,265,189 FRAX and 21,462 OHM.

Using the RFV calculation for v2, we get an RFV of $891,851.

V3 Position: FRAX LP position from current RFV/OHM of ~$25 to $1500.

We move our 9,265,189 FRAX and 21,462 OHM into a Uniswap v3 position with a range of $25 to $1500, assuming a price of ~$431. Using the RFV calculation for v3, we get an RFV of $127,030. Overall this is a 1.9% decrease of our total treasury RFV.

Capital Efficiency

In this example, the efficiency of the position more than doubles, but the RFV of the position goes down compared to the same deposit in v2 liquidity.

In a v2 position, the protocol will continue to spend its FRAX until OHM hits $1; it will accumulate quite a bit of OHM between $25 and $1. By contrast, it will have spent all its FRAX on OHM in its v3 position by $25. This is the cost of dramatically increasing capital efficiency.

While the calculated RFV of the position goes down, the capital efficiency of the position increases dramatically. In practice, increasing capital efficiency means that traders have to sell (or buy) more to bring the price down (or up) by the same amount as in a v2 position. Consequently, slippage goes down by 50% per trade every time capital efficiency doubles. A further benefit is that the protocol captures twice as many fees per $1 of volatility every time capital efficiency doubles.

Overall, the tradeoff of RFV for capital efficiency is a balance that must be struck between risk and reward; we believe that implementing a v3 pool between our current RFV per OHM and ATH price of OHM represents a solid balance of risk and reward.

Vote

(1) For: Move the OHM-FRAX pool to Uniswap v3 with the proposed parameters AND move up to 50% of this pool to Arbitrum when FRAX moves their liquidity

(2) For: Move the OHM-FRAX pool to Uniswap v3 with the proposed parameters, BUT DON’T move up to 50% of this pool to Arbitrum when FRAX moves their liquidity

(3) Against: Don’t move OHM-FRAX pool to Uniswap v3 with the proposed parameters AND DON’T move 50% of this pool to Arbitrum

Informal poll

    Would there be utility in retaining OHM-FRAX Uniswap V2 bonds, with the intention of migrating that captured liquidity into Uniswap V3 over time? The intention here being to keep the portion of liquidity owned by the protocol as high as possible.

      This sounds like a very exciting opportunity and significant strategic move forward.

      I've read the proposal and the math deep dive, and I'm still a little fuzzy about one particular scenario.

      Question: In the hypothetical "Bank Run" scenario, what market behavior might we expect to see between price points $25, when the liquidity is exhausted, and $1, when the protocol begins to buy and burn Ohm?

        Safisynai Right now no, as the capacity for FRAX bonds is very low anyway, we've pretty much reached our targets for now. We're not against other LPs, we're just not incentivizing them.

        plutus We still have liquidity on our main OHM-DAI pool and we can easily adjust this liquidity range on v3 if needed.

        For: Experiment with v3 liquidity. Team will gain a lot of experience I believe. This could be valuable to document in some kind of lessons learned. In case in a distant future there needs to be handover from old to new team members. This way new team members will be able to stand on the shoulders of giants, rather than to start anew.

        Against: The amount of capital that Olympus puts to risk on the Arbitrum layer should not only be contingent on what others do, but rather what the state of Arbitrum itself is in. Is it still alpha? Are the smart contracts still on single private key? This should perhaps have more weight than decisions from the FRAX team.

          Reads like a smart move, particularly on adding liquidity to arbitrum. New market to take share in. The efficiency in fees as stated is also material to suggest the approval. Well done team.

          bubbidubb Your point about smart contracts is well-taken. I'll be interested to hear the response.

          Uni v3's mechanism is a fantastic innovation that allows market participants to achieve much higher levels of capital efficiency. I'm fully aligned with the proposal seeing as we're looking to achieve just that; thicker liquidity, lower slippage, and higher revenue from trading fees per unit of capital are some of the consequences of higher capital efficiency of the protocol owned liquidity.

          The apparent downside is that the RFV goes down, though the proportion of the reduction doesn't convey the amount of risk that is taken with this sort of decision -- if that could be called 'risk' at all. The trading ranges are adjustable, and that goes both ways! 🙂

          Really looking forward to the new opportunities the Treasury can take which are both safe and effective.

            bubbidubb We are waiting for FRAX to move so that there are more FRAX pairs users can move through, not just OHM-FRAX. That way we get maximum utility out of the pool.

            I appreciate the proposal and the idea to make our liquidity more capital efficient. I would just think that fragmenting our liquidity by only moving 50% to Arbitrum doesn't make much sense, because the other half will remain useless. So I would rather vote to move everything we have to V3 or Trident and if it makes sense have all of it on Arbitrum.

            vira The apparent downside is that the RFV goes down

            I imagine that in the longer term, if we needed to, instead of managing the Uniswap V3 position itself and holding the hard-to-value position NFT, we could make use of third party Uniswap V3 management partners where treasury ends up with an easier to value, fungible token, there's already a bunch of these starting to pop up (Visor, Charm, Sorbet, and I'm sure at least a few others).

            • vira replied to this.

              Safisynai

              Yup! Lots of opportunities down the line, but we have to be sure we've properly evaluated the stacking protocol risks when implementing 3rd party strategies.

              So you checked projects like visor finance and it wasn't worth the risk?

              I notice that the is a OHM-ETH pool on visor with:
              TVL USD

              $622,386.50

                bubbidubb Well in this case, we'd be earning less fees, but we'd also get "free" (in terms of us not paying out incentives) liquidity. I think that's a good deal for us, as the payment to those liquidity providers could be considered a rev-share scheme.

                However, keep in mind that there already is a pool on Uni v3 not controlled by us earning fees as you can see in the proposal, so what you're saying is already happening, just without us there.

                This is a fantastic idea. Using V3, we can also layer several ranges of liquidity. For example considerable liquidity (something like ?%) from 25 to 1500, and then extra liquidity from more recent lows to highs 150 to 1000 (25%).

                Barring major events, I don't see Ohm trading below 160 anytime soon.

                I think this could be a good idea, but I'd like to understand the impacts of ending the "OHM-FRAX bond". I feel like ending the OHM-FRAX bond and migrating 50% of the OHM-FRAX pool to Arbitrum should be separate OIPs and can't be bundled together so people can vote with on single things rather than a bundle of things.

                  bubbidubb I agree with this.

                  I think Arbitrum is super important for investors like me that can't afford the ETH gas like myself, but I also want to be sure that Arbitrum isn't putting liquidity at risk since it's relatively new.

                  I don't know enough about this, but before migrating too much liquidity into Arbitrum I think we should make sure the Arbitrum contracts and even the Olympus web app/source code get audited to make sure there is less risk.

                  SIDE NOTE: I strongly recommend we get OHM on defi insurance platforms: https://forum.olympusdao.finance/d/93-defi-insurance-offerings-for-ohm because even with audits and great coders there's always risk. Insurance doesn't stop hacks, but it protects funds which is the next best thing.

                  If folks are opposed to using one of those platforms we can also use something similar to AAVE's safety module which locks up staked funds to protect user funds in the event of an emergency https://docs.aave.com/aavenomics/safety-module.

                  i-feel-so-al-ohm They are separate options in the vote, so you don't have to vote for both. Olympus on Arbitrum already has a separate proposal where the pros and cons are discussed.

                    Graz likely one step at a time - this is the passive management option to get us going. I think the team will continue to look for active management strategies!