BIP 74. Treasury Controlled Liquidity for Badger

[Mr_Po]

Category: Treasury management.

Scope: Uniswap V3 Treasury Controlled Liquidity strategy.

TL;DR:

• Establish Treasury Controlled Liquidity on Uniswap V3 to cover the (0; + infinity) range of Badger/WBTC price.
• At 0.0005 Badger/WBTC price, the model would require 254.85 BTC to cover an 87.5% price reduction and 1,082,734 Badger for a 32x increase.
• BTC from price ranges significantly below the current price would be available for productive use.
• Incentivization of Uniswap liquidity would stop once TCL is established.

Overview.

BadgerDAO has accumulated enough assets in its treasury where we have an opportunity to provide a similar liquidity depth with treasury controlled liquidity (TCL) with Uniswap V3 as we have now with incentivized liquidity on Uni V2 + Sushi.

The main benefits of treasury controlled liquidity are that it is sustainable, reliable, and can scale better, plus the DAO would be earning trading fees on top of it.

Utilizing Uniswap V3 will allow making that liquidity more efficient:

1. With Uni V2 type of liquidity, half of BTC on the buying side of LP covers the (-100%; -75%) price range. With Uni V3 TCL, we’ll be able to utilize those funds productively and pull them back when needed.
2. With a set of Uni V3 positions, we can increase the liquidity depth at a higher pace than Uni V2 in the scenario where Badger price appreciates.

With the current emission schedule, the DAO spends 8,926 Badger per week for 28 $M liquidity on Sushiswap and 12 $M liquidity on Uniswap. Establishing TCL would effectively double that liquidity. When it happens, the plan is to stop incentivizing Uniswap liquidity in line with BIP 69 and thus spend twice less.

Specifics.

Here’s how the proposed core position would look:

In the table above, the liquidity is split into multiple NFTs on Uniswap V3 that are attached to specific price ranges.

So at 0.00050 Badger/WBTC price, there would be 115.47 BTC liquidity from the current price to -50% of the current price (in the (-50%; 0%) price range) and 81.65 BTC in the (-75%;-50%) price range.

If the price moves up to 0.00100, there would be 200 BTC in the new (-50%; 0%) price range (with the 282,843 BADGER originally deployed in the range having been sold for BTC within the LP). You can check out how the TCL model would scale on the upside compared to UNI V2 type of liquidity in this sheet.

Managing the Liquidity

The long-term goal with managing the liquidity is to have at least (-50%; 0%) price range covered at all times.

For example, if we started at 0.00050 Badger/WBTC price, and the price moved down to 0.00025, it would activate the 0.000063-0.000125 position with 57.74 BTC, which would be pulled from the dogfooding allocation (BIP 65). Then, if the price would move back up to 0.000375 (the middle of the price range), the lower price NFT would become available for dogfooding again.

In the beginning, though, before the automation infrastructure is in place, I would go with a more conservative management structure, where at least (-67%; 0%) price is covered at all times, so the price triggers for the events would be higher. So in the example above, 0.000375 price would trigger the 0.000063-0.000125 position deployment, and 0.00050 price would allow for pulling it back.

As for positions above the current price, there is no particular need to actively manage them since there are ample amounts of BADGER in the treasury, so a wider range could be established initially.

The suggestion is to cover the price range up to 0.0016 BTC and approve adding more if the price moves above 0.004 (~$240 BADGER @ $60,000 BTC).

Implementation.

Establishing the initial position

The exact Badger + BTC requirements from the treasury will depend on the price at which TCL gets established and future Badger/WBTC price behavior.

Deploying TCL at a lower price would require more Badger and less BTC, deploying at a higher price would mean the opposite.

Given that a limited amount of BTC is available in the treasury, deploying noticeably above 0.00050 would mean that we’d need to make the BTC positions below 0.00050 a bit “thinner.”

For example, if we were to establish TCL at 0.0004345, we’d deploy

• 57.74 BTC in 0.000063-0.000125 range
• 81.85 BTC in 0.000125-0.00025 range
• 56446 Badger + 89.17 BTC in 0.00025-0.0005 range
• 1,082,734 Badger in 0.00050-0.01600 range

And if the price of deployment is 0.00065, we’d need 67.69 BTC + 163,980 Badger for the 0.00050 - 0.00100 range. At that point, we’d need to take those 67.69 BTC from the virtual allocation of the lower ranges and establish a slightly different structure there.

Even though the BIP would unlock TCL for the (0; + infinity) range, most price ranges wouldn’t get activated - and thus, there wouldn’t be the need to pull BTC to cover the lowest ranges or Badger to cover the highest.

Establish Treasury Controlled Liquidity for Badger according to the model above.

• Yes
• No

[Tritium]

This BIP would be stronger with a timeline, or more implementation details. I voted for the concept, but a few questions about implementation.

1: Do we really need to pull 1.6 million badger from the treasury and out of circulation now. Can we wait to fill out much higher ranges until we get closer to it?
2: To that note, would be nice to know how much you feel needs to be TCL’d this year, and what would be left next year.

That being said the BIP is important to get done soon, and this gives leeway to establish any timeline so I’ll vote for. The BIP would be stronger/have less questions around it in the future, if at least the first implementation steps were laid out in as much detail as possible given the changing market.

[Devin]

I like this proposal, it’s more capital efficient and sustainable then continuing to pay for liquididty on uni V2. How soon after implementation would rewards in uni V2 be discontinued?

[Mr_Po]

The timeline depends on the execution from the tech, which will be done when it’s good to go. I don’t see much value in putting random numbers there like “it will be done in one week”.

1. In reality we wouldn’t put 1.6 million Badger into circulation, as it’s all conditional.

If we establish TCL at 0.0005, and Badger price goes up to 0.0010, then there would be 283k Badger added to circulation - or bought from the treasury.
If the price goes down to 0.00025, then 327k Badger would be pulled out of circulation, or bought by Badger treasury.

So the fact that there would be some Badger sitting in the 0.008 - 0.016 range doesn’t really change anything.

2. From the econ side, I see no reason to not do all the TCL allocation right after the BIP gets approved by the governance, the pacing would 100% depend on the tech / multisig operations.

[Mr_Po]

I gather from Tritium’s comments that the implementation will be step-by-step, with smaller chunks of funds deployed at a time. To me discontinuing Uniswap incentives after we have similar liquidity depth on Uni V3 sounds reasonable, as it would mean no drop in available liquidity for the token.

[cryptomooniac]

Read this in detail. Totally for it.

[neekolas]

Excellent proposal. I vote Yes!

[imimim]

Isn’t Popsicle Finance, which is a friend to the badger dao, building out a product to manage Uni V3 liquidity like this? Has utilizing their service, or a similar one, been explored? What would the advantages/disadvantages be?

[Mr_Po]

Popsicle product wouldn’t manage the liquidity like this for BadgerDAO, and yeah, we have explored the options. There’s been an extensive discussion about it in the rff channel on Discord.

In short, these products are fee optimizing, not liquidity optimizing, and they imply value loss due to rebalancing.
And we don’t want TCL to rebalance, to buy high and sell low, most importantly we don’t want our treasury to sell its Badger when the price goes down because of that rebalance.

We want deep and reliable liquidity, so even risks aside, the concentrated liquidity rebalancing model isn’t that suitable for TCL in my opinion, at least not for the core position.

If we wanted to maximize the fees, we could just cancel incentivized liquidity alltogether and capture the majority of the fees from trading volume in TCL. And at that point the fees that we wouldn’t capture would be worth it too, as it would mean that someone provides liquidity to Badger “for free”.

With the TCL model we’re trying to achieve deep support on the “buying” side of the LP.
This support would grow further/faster in the scenario of Badger price appreciation - and at the same time it would unlock more funds to utilize productively, while these funds would still become available on the buying side of LP if needed, without selling Badger.