ARCx Uniswap v3 pool

Problem

The intention of this post is to offer a concentrated Uniswap V3 position as a solution to liquidity problems we have been having. ARCX continues to support and subsidize sushi LP (SLP) with Farm 9, but the pool has yet to reach a critical mass that has significant depth.
Right now, the +/-2% pool depth on the sushi WETH/ARCX pool (the maximum transaction size at which 2% slippage occurs) is approximately $13k USD or 5 ETH. This means any individual wishing to buy ARCX in larger allotments may be deterred by significant slippage, and at even just 15 ETH, a user receives a ‘swap anyway.’
Additionally, low liquidity leaves the ARCX token unattractive to market researchers from a fundamental perspective and very susceptible to market manipulation.

Solution

First of all—and this is completely aside from the ‘problem’ as stated above, but no less important—Uniwswap is the 69% of the DEX market by volume. We need a Uniswap pool from a market integration perspective. Many use DEX aggregators, but many users still navigate to (and do market research on) Uniswap directly.
Returning to the ‘problem’ as stated above is where Uniswap v3 comes into play. The capabilities of v3 are beyond the scope of this post (for more reading, see the official recommended doc page), but the major strength is that a concentrated positions (positions with lower and upper limits) can:

  1. Require much (up to x100s) less liquidity to achieve similar depth and efficiency as a full-range pool
  2. Be strategically placed to support buy-side pressure, rather than sell-side pressure

Proposal

There are endless debates and simulations to be made regarding Uniswap v3 positions. To make a long story short, I want to propose that ARCx DAO use some amount of the treasury of ARCX tokens in a concentrated Uniswap v3 position just above the current market price and to several multiples of that current market price. By placing the position just above the current market price, it will only require a one-sided contribution of ARCX, and will support buy-side orders going up from the price on the day of deployment onwards.
I’m very willing to discuss other formulations, features, or specific ranges that others find valid, but there are at least 2 ranges on Uniswap v3 that would better serve market orders than currently capable from the sushiswap pool.
For reference, a full range v3 position (equivalent to a v2 position) is listed first here; without further ado, the options I have calculated for a ~$50,000/25 ETH swap to have 2% slippage or less (4x the current depth capability of the current sushi pool):

  1. [Full range], $2.5 million liquidity required
  2. [Market price - 6x market price], ~$830k liquidity required (3x more efficient)
  3. [Market price - 3x market price], ~$500k liquidity required (5x more efficient)

How these pools would work is (assuming ARCX/ETH pool):

  1. Sell $1.25 million ARCX for ETH to acheive 50:50 balance for full range
  2. Deposit only ~$833k ARCX just above market price; as ARCX price increases, it gets converted to ETH, up to the point where it acheives x6 it’s current market price (vs. ETH) and then the position would be 100% ETH
  3. Same as 2, but deposit is $500k ARCX and is converted to 100% ETH at x3 it’s current market price (vs. ETH)

Note: 6x the current market price (approximately $2.7 USD/ARCX) is about as high as the ARCX token has ever gone (pre-split or otherwise); 3x the current market price is about is higher than the post-split token has ever gone

The above positions 2. and 3. would also require some kind of direct oversight to update should when the buy-side pressure goes strongly above the range in the future. My recommendation would be that when the pool regains a 50:50 weight (say at 3x current market price in option 2. for example), the pool be re-evaluated by the treasurer team or in community call or…?

ARCX/ETH pair is proposed vs USDC because (a) it’s the most popular pair, again aligned with liquidity on the most popular DEX, and (b) like the ‘just above market price range’ feature, we’re not trying to explicitly facilitate exit liquidity (which usually wants to go directly to stables).

This topic is tagged as a discussion, but I want to move it forward to a formal proposal ASAP if it is to use ARCx DAO treasury to seed this position. My preference is for option 2. above or a range like it. Please weigh in below. Thanks!

7 Likes

This seems well timed to me and I’d be supportive of either options 2 or 3. If I had to choose, I’d lean towards the 6x because of market conditions (appears that ETH and DeFi are strong), hype ahead of the DeFi Passport release, and discussions around token utility.

Thanks @monadnoc

I support this, great writeup @monadnoc.

This would certainly increase price discovery of the token and ARCx suffers a lot from poor awareness and accessibility today. As you mentioned, it will also be helpful to get on aggregators like 1inch.

I agree with @acedabook that options 2-3 seem optimal. Llama DAO can help monitor the LP position. Also agree with using the ARCX/ETH pair.

As far as the treasury position goes, we currently have 6.1m ARCx + 7.1m USDC (total value $10.1m). Here’s what options 2-3 would look like:

One question: which fee tier on Uniswap will we use for this? I’m guessing either 0.30% or 1%.

3 Likes

Solid write up @monadnoc. I’m thinking we can probably execute option 2 but through financing it in a different way:

  1. Rather than selling ARCx for ETH we instead just use the USDC in the treasury to purchase ETH
  2. Given this is for the community and token rather than to finance operations directly we can just mint the ARCx directly (especially since we’re not close to the first year limit of 100m tokens).

@acedabook’s point about ETH and DeFi being potentially explosive should make us probably optimise for the 6x range especially if we’re going to mandating ARCx to be used as deposits for the DeFi Passport tbh.

So in summary we’d be:

  1. Purchasing ~$800k of ETH via USDC in the treasury
  2. Minting new ARCx tokens that would provide liquidity alongside the purchased ETH
  3. As an aside, redeeming the existing 1inch/ARCx LP shares and then creating a new 1inch/ARCx pair with the post-split token

Curious to hear everyone’s thoughts on this.

3 Likes

hey guys, several questions/comments from me.

  • Placing your liquidity above the current price ratio only addresses liquidity issues when the ARCX token price appreciates. With a wide range like from just above the current ratio to 6x, the benefits of option 2 being 3x more efficient and option 3 being 5x more efficient won’t be achieved until you hit the middle of the range. So you are basically not providing any liquidity at current price + the full benefits of concentrated liquidity will unlock only when the price approaches middle of the range. Is that what you want?

This is an important parameter and I think this should be a 1% pool.

You actually don’t need any ETH for option 2 or 3, as you provide single-sided liquidity above the current ratio.

I would like to suggest a slightly different variation of the proposed options - provide X amount of ETH-ARCX liquidity across the entire range (same as V2) + provide X amount of concentrated liquidity around the current ratio which would be actively managed.

The first 2 steps will be the same.

You now have $800k of ETH and $800k of ARCX. You provide say $1m of liquidity across the entire range and concentrate the remainder $600k around the current ratio. Another interesting configuration for the second part is you stagger your concentrated liquidity meaning you create several different LP positions with different ranges. So you provide some concentrated liquidity +/- 20% from current ratio, add more from +20% to +80%, etc.

This would allow there to always be some base level of liquidity + some concentrated liquidity to give you the capital efficiency benefits.

4 Likes

Following a tweet, I saw this discussion earlier and made an account to come back to comment.

However, I think @verto0912 has covered most points I was going to make

I would possibly disagree on this point. I think as soon as the price moves into the range you will see benefit ($X,000 trade to move the price by y%). However, when you are only just in the range, once you hit the limit, there is a cliff edge to the liquidity. i.e. Deep, Deep, Deep, Deep, nothing


Allowing buy-side liquidity only has the benefit that you allow people to buy into ARCx (but they become stuck). It does, however, allow more volatility on the downside.

Multiple, overlapping v3 positions makes sense to me. They are easy to monitor and you can pull them at different times (so remove 100% ETH as the ARCx:ETH ratio increases).

1 Like

1. Deciding a fee

I agree due to scarcity of ARCX/ETH LP and risk to treasury in funding it, and I have opened a test pool with 1% fee (see below)
2. Deciding a price range

I wasn’t sure about your claim, @verto0912 so I set lofty goals to figure out the math behind v3 liquidity distribution and efficiencies–and I failed to devote enough time to come up with a exact formula or analytical solution. At the risk of exposing my laziness and my burner ARCX holding account, and since gas prices were so cheap this weekend, I just opened up a 1% fee ETH/ARCX pool.
The results and conditions for the test pool were:

  • 1% ETH and 99% ARCX in the [market price]-[3x market price] listed above (option 2)—should have been ‘5x more efficient’ based on v3 position simulator.
  • Total deposit was ~$14k—no this isn’t chump change to me. I just didn’t have this ARCX in Farm 9 and I wanted to put it to work, and I wanted to make sure the pool was large enough to easily get some authentic rebalancing compared to the sushi pool
  • 2% of $14k is ~$280, but the 2% price impact level was actually 0.32 ETH (~$650 as of right now), when it rebalanced to 10% ETH and 90% ARCX, price 2% price impact level only increased to 0.33 ETH (~$670 as of rn)—so about 3x more efficient than full range pool (even though we wanted 5x in this range) and this efficiency must indeed be concentrated smoothly around the 50%:50% price range, but I don’t have a clue the exact distribution (normal/gaussian?) for an analytical solution
  • as of right now, the v3 pool is not necessarily showing up on wallet-integrated DEX aggregators (Argent is the only one I tried, for example); nor does it have enough data or trade history to show up on coingecko to confirm the 2% depth

So the test pool tells us that @verto0912 is right that the edge of the pool is less efficient, but it’s still efficient; we just need to take the edge case diminished returns in efficiency into account. (Aside: the 1% fee range works as expected)

3. Needing multiple pools or not

As of right now, the Sushiswap pool will continue to incentivized and work as our full-range pool, so just by introducing this new v3 position, we have both a concentrated position and wide-range position.

However, I understand the need to micro-manage and allow for some sophistication with pool management (see below)

4. Revised proposal with multi-stage, multi-pool approach

@verto0912’s point is confirmed by the test pool. Considering all of the input and the lessons learned from the test pool, I would like to propose the following:

  1. Keep SLP as full-range supplement for liquidity (at least as long as Farm 9 is running)
  2. Mint $800k ARCX for single-sided deposit in [1% above market price]-[3x-market price] range—as small of an increment out of range above current market price as possible with a 1% fee uniswap pool. (Note: pool already created and origination fees in ETH already paid); based on 3x efficiency as noted from the test pool, this should give us liquidity at the LP position edge to enable ~25 ETH purchase of ARCX with 2% price impact or less.
  3. This pool in (2) will be formally controlled by @schlabach and/or colleagues at Llama
  4. The pool in step (2) will be monitored daily—if I take a leave of absence from monitoring it, I will post in the discord who is to cover for me; if it remains within less than 5% (or any % greater than) the 50%:50% ETH/ARCX distribution, at least half the pool liquidity will be removed and placed at a [1% above HIGHER market price]-[3x-HIGHER market price] range—this actually exceeds the price growth expectation proposed in my original 3rd option, seconded by @kerman and @acedabook and aligns with a multi-pool strategy requested by @OverAnalyser and @verto0912
  5. No later than Q3 of 2021, the Uni V3 pool position will be summarized for community, reviewed, and either kept as is or readjusted for more complex strategies (or incentivized for community members to take over these LP ranges)

If the above is palatable and seems transparent enough to all, a formal AIP is the next step.

3 Likes

Great overview, thanks for that.
With this in mind, and thinking ahead towards the potential end of the SLP incentives. What would be the disadvantages of going down the road of our friends at Olympus DAO and owning the majority of the liquidity? And thus saving incentives. I’ve mentioned this in the discord but I wanted to get it up here too.
At this early stage in the project where the choice of DEX is in our own hands, it could be a great way to earn, as any trades are fees going back to the DAO.

Absolutely this could open up the potential for the DAO to manage all (or at least a huge chunk) of its liquidity.

A quick update:
(1) Since ARCX tokens have already been allocated for liquidity incentives, we likely won’t need a formal AIP
(2) If the initial concentrated position does not go into range within the first 7 days, I propose that it should be removed and reset at whatever new market price sets it closer to the ETH/ARCX market price (so that it can get into range quickly again)
(3) If desired, if too much of the ARCX has been converted to ETH with this procedure, a very concentrated position just below a strategic price could be deployed as a ‘buy-wall’ to protect and buffer that price, for example (as originally described by https://twitter.com/CometShock/status/1412597672809320450?s=20)

The above discussion has been executed successfully. Key txs:

  1. Minting of 1.6m (1m for LP + 666k to DAO emissions distributor) ARCx - Ethereum Transaction Hash (Txhash) Details | Etherscan

  2. 666k sent to emissions distributor: Ethereum Transaction Hash (Txhash) Details | Etherscan

  3. Uniswap v3 LP position created with a range from the current price ($0.84) - 3x current price: Ethereum Transaction Hash (Txhash) Details | Etherscan

  4. Uniswap v3 LP position NFT transferred to AGA managed by ARCx and Llama: Ethereum Transaction Hash (Txhash) Details | Etherscan

Thanks to everyone here for getting this over the line :slight_smile:

2 Likes

Hi everyone, I’m Michael from Llama. @schlabach filled me in a few days ago to help monitor this, glad to see that the LP position just went up!

I’ve been working with Uniswap v3 data for a bit and just put up this dashboard up to help track pools and LP positions.

Dashboard link (pre-filtered to the ARCx/ETH 1% Uni v3 pool - zoomed in on trailing 3 days): Dune Analytics
Screenshot:

In this screenshot, you can see the increase in liquidity added, trading volume, and LP composition. In the dashboard you’ll also see share of volume and TVL by DEX (i.e. Uni v3 vs Sushi). It’s still a work in progress, the last big piece I want to have in by today/tomorrow is tracking price impact of trades over time, which I know is super important here.

Another callout, is that the chart on top right is showing price as “# of ARCx per 1 ETH.” So, as ARCx appreciates in relation to ETH, the price should move to the left.

Definitely let me know if you have any other feedback/ideas on what could be useful. We should have some early insights soon.

4 Likes

Hi all, as an FYI, the position is now out of range since ETH has had its run.

1 Like

Yup, spoke with Daniel about this today. We need to readjust the range to be a bit lower than what it is now.

@kerman @monadnoc

We invest directly in ArcX through an Australian based crypto fund, Mt Buller Labs. You may have seen Al McCann on some of the community calls each Tuesday?

I thought this proposal from the Lido community may be useful to help in managing our ArcX Uni V3 positions. They are working on a partnership with Visor to increase the Uni V3 yield for the Lido DAO.

Could be worth considering for ArcX to reduce the management overhead for the ArcX position and likely increase overall yields.

Thanks for posting this. We’ve been discussing and will continue to monitor. Right now it doesn’t look like it’s worth paying the 10% management fees. We already have the Llama team managing the treasury so it makes sense to have them working on the Uniswap v3 position too.

I’ve just put up a new AIP to boost liquidity, would appreciate you guys input. I realise it’s short notice.