[Proposal] Balancer V2 Liquidity Mining Program

First off, let me fully put my humble weight behind this suggestion.

Next, I feel the need to ad some points on my own.

There seems to be a huge consensus around the idea that we need to incentivise LPs with rewards. Returns are made out of rewards plus fees, and so the consensus implies that Balancer is incapable of generating enough fees to attract LPs. I disagree with this notion. More focus could be put on attracting volume and reducing other risks for LPs. Paying LPs for participation is not a sustainable model. Using scarce resources like BAL to invest in unsustainable advantages is not good practice.

As I feel alone in the above sentiment, I don’t expect the community to agree and so given that we have to incentivise LPs, I strongly believe they should support a broad and diverse base of tokens - as this plays into the unique properties of Balancer. Incentivising a few pools centered around a few core tokens feels to me like a “we also want to be Uniswap” strategy, which again, is not good practice. Embrace what makes Balancer unique, do not use BAL rewards to undermine it.

Why are we not considering incentivising tokens instead? As I understand V2, the notion of “pools” is less of a fact, and more of a concept and so this could easily be reflected in rewards too. Whitelisted tokens (and I do believe some QA is warranted for tokens to be eligible) could share the reward pool and be capped for rewards at some sensible level derived from trading efficiency. This also allows for special treatment of BAL, who could be the only token in the superior tier. This doesn’t really require the ballers to do anything but white- and blacklisting tokens, but it would make for a much more organic and natural ecosystem.

Balancers strengths are in flexibility and diversity. We should make that efficient and safe (those are two very different vectors), as this will sustain an ecosystem that is responsive to signals in the market and also attractive to a significant number of LPs, aggregators, arbs and end users. I think lower eth fees is a big win and a major contributor to increased volume. In addition, parts of those saved tx costs can clearly be captured as fees as well.

Let’s be a bit smarter about this than just providing more BAL for BAL holders. I get that it feels good, but seriously. It’s bad practice.

3 Likes

I think that on a long term basis everyone agrees, but in the short term the liquidity mining rewards doubtlessly attract LPs, which is essential for being able to facilitate larger trades and compete on volume with other DEXs.

My understanding is that the intention isn’t to run rewards forever, but to get Balancer to a stage where it’s the primary source of programmable liquidity on ethereum, so that trading fees alone make pools profitable. Rewards will be voted out by governance when they’re deemed no longer necessary.

It’s the old two-sided network problem: you need trading volume to attract LPs, and LPs to attract trading volume. Rewards short-circuit building one side of the network by offering APY to LPs in the short term, so that Balancer can grow to a critical mass where they’re no longer necessary.

3 Likes

I strongly agree with these two ideas.

1 Like

Another thought here in regards to distributing BAL rewards is to slowly lower the amount of rewards given to major pool receivers in relation to the price performance of BAL token. A flat percentage APY could be maintained, ideally, rather than a flat rate of BAL distributed. This would be very fair and allow for further BAL distribution to other pools without making V1 long term BPT holders suffer. And for now we really ought to keep the whales right where they’re at. They’ve taken the most risk to be here and they provide the most liquidity to the platform.

P.S. - As a smooth brain speaker I am awed by everyone here. I apologize in advance if I missed someone’s thoughts that may have already addressed this idea.

“You” in this post refers to “all of you guys who seems to disagree with me”, not focused on @bakamoto20 in particular :wink:

You think we need to incentivise LPs to come around. I think you are right for a lot of liquidity, but I’m, not convinced;

  • we know what share of liquidity depends on rewards, what share of liquidity is removed if we cut rewards in half?
  • we know how much liquidity is considered critical mass
  • that the proposed incentive system takes either of the two preceding points into account

You think the best way to deploy incentives it to focus them on select pairs or tokens, I think if we have to use rewards they should be much broader distributed in support of the whole ecosystem. If we focus on select pairs for deeper liquidity, we are trying to become more like Uniswap. If we go the other way, we become more “Balancer”. I am very much of the opinion that providing liquidity to a lot (no, a LOT) of tokens is the right way to go for balancer (yes, just opinion, not data)

Based on the suggested rewards structure, You think that the more liquidity the better - for the select pairs or tokens.

  • I have not seen any data driven opinion about the marginal usefulness of added liquidity. There very likely is a right answer to this, but “as much as possible” is not it.
  • when we incentivise more liquidity on the platform than it can use in a valuable way, we take down the rewards for the existing liquidity. That is, we introduce inefficiencies in the economic model that we then have to compensate for by providing rewards. And we do this without knowing the answer to the previous bullet!

Finally, on BAL. It is clearly of high value to the project and should be used accordingly. When we release more BAL we also dilute the future value, whatever it may be. This dilution, the risk many talk about, would be far less if we stopped giving it out in the reckless way we now do. Again, ceasing the practice, would remove or reduce the need for it. Granted, BAL needs to have utility beyond accumulating intrest in it’s own currency and platform governance - but that’s true anyway. To me, Stacks seems to be on to a very interesting idea. Stakers are paid not in STX, but in BTC. This removes the inflationary pressure and turns STX into cashflow instrument.

I don’t have the time, tools or data available to do the calculations to find out what options we have and the best decisions we can do on them, but I do not see traces of a lot of data in the proposal and above comments, and that’s … bad practice.

Much appreciated for everyone’s feedback!

This has been a healthy discussion and we hope to address everyone’s concerns. From our perspective, there were two main points of pushback:

  • Yields for BAL LPs under this proposal will be too low
  • Ballers have too much control over the liquidity mining.

After discussing internally along with some of the folks at Balancer Labs, we want to amend the original proposal to address the above concerns. Given that the proposal already has enough voting participation on the poll in favor “for” (70% in favor, 30% against), **the proposal outlined below will move to a formal Snapshot vote on Friday, April 15th in order to prepare for the upcoming launch of V2.

We hope that everyone participates and weighs in their sentiment on-chain.

We’d also like to highlight that no members of FireEyes DAO will participate in the committee as a result of the feedback.

I’d also like to highlight, as mentioned on Bakomoto’s feedback, that this system can always be changed in the future via normal governance processes if the community deems it unfit after its implementation (assuming it passes). That’s the beauty of open, decentralized governance!

Lastly, the section surrounding the govFactor has been removed and made into its own proposal. You can view, comment, and vote on it here.

Here’s the official proposal that will be featured on Snapshot:


Motivation

Crypto moves fast and the AMM space is as competitive as ever. In order to stay relevant and to attract significant liquidity, Balancer needs to implement a flexible Liquidity Mining Program which targets high priority pools and can respond quickly to changing market conditions.

Unfortunately, this can result in significant overhead for normal BAL governance. Constantly weighing in, voting, and checking the forums on a daily basis can be a hassle for most BAL holders.

As a result, this proposes having a dedicated committee of core members to the Balancer protocol to lead this charge and ensure that the protocol is always incentivizing the right pools, at the right time.

Specification

Balancer V2 Liquidity mining incentives will operate in three tiers governed by a combination of BAL governance and the Ballers Committee.

Tier 1 will be voted on by BAL governance while Tier 2 and Tier 3 are voted on by the Ballers committee.

All Balancer community members will retain the right to make proposals and signal support for T2 & T3 Liquidity Mining inclusion. The Ballers Committe will act as a liason between community sentiment and formal changes to the program. Discussions on Liquidity Mining inclusion will be open for all community members on both Discord and the Forum.

The Ballers Committee will have no control over the Tier 1 slots. Instead, they will be governed by normal BAL governance which will be put up for vote every 12 weeks. Additionally, all decisons made by the Ballers Committee (along with voting addresses) will be made public for everyone to see prior to implementation.

In order to accomodate for sufficient yield to attract liquidity on certain pools, more than one slot may be allocated to a single pool by consuming additional T2 and T3 slots as voted on by the committee.

The timing for changes on each tier outlined above is a guideline–BAL governance and the Ballers Committee will always have the right to change tiers if the market conditions deem it necessary (examples include rug pulls, bugs, and other issues which would require quick response by governance).

BAL/ETH Pool

In order to mitigate the reduction in yields for the BAL/ETH pair, the Ballers committee will aim to target the existing returns in V1 (60-70% APY) through the use of additional T2 & T3 slots.

It’s important to highlight that there are a significant amount of variables that will affect the BAL/ETH APY (as well as all other pools). This includes amount of liquidity migrated during the transition, amount staked, tiers consumed, etc.

With all those variables, The Ballers Committee will be responsible for trying to ensure that the APY on the BAL/ETH pool are in line with the above.

Transition from V1 to V2 Mining

The transition from Balancer V1 liquidity mining to V2 will be gradual and take several weeks. This will start with only a few of the Tier 3 (T3) and Tier 2 (T2) slots being filled by top priority pools, with new slots being "activated’’ on a rolling basis, targeting every week.

At every activation, the corresponding amount of BAL is subtracted from the weekly amount available in the V1 liquidity mining program. This way the total BAL distributed through liquidity mining between both V1 and V2 pools will always stay the same at 145,000 BAL per week.

As the community becomes more confident that there are no vulnerabilities in the V2 smart contracts, more of the liquidity mining program is expected to shift from V1 to V2 until the entirety of the V1 allocation is exhausted. This can happen by having more of the available slots being assigned to V2 pools, and also by upgrading a V2 pool to a higher tier.

The transition period will aim to take roughly 8 weeks from the start of V2 mining. At the end of the transition period, Tier 1 slots will be allocated to these pools:

  • 80/20 BAL/ETH, dynamic fee
  • 60/40 ETH/DAI, dynamic fee
  • 50/50 ETH/WBTC, dynamic fee
  • DAI/USDC/USDT stable pool

Ballers Committe

The selection of Tier 2 and Tier 3 Liquidity Mining Pools will discussed on a weekly basis according to community input and signal, and voted on formally by the Ballers Commmitee.

The Ballers Committe is open to everyone. Any engaged individuals in the community can be elected by fellow community members to join the committee. More on this below.

The goal of using the Ballers Committee is to avoid extensive overhead on BAL governance as well as to provide quick responses according to market conditions.

Ballers will use a specific Snapshot space (separate space from traditional BAL governance) so that the community can easily keep track of those decisions.

All decisions made by Ballers will be made public on Balancer’s forum shortly after the end of the committee weekly meetings (ideally within 2-3 hours). Highly respected individuals and working groups in the space may also be invited by Ballers to these meetings to provide strategic input. That said, they will have no formal voting rights on Snapshot.

Additional Background on Ballers

Ballers are active members of the community that have gone above and beyond by contributing significant resources and time to the Balancer Protocol. There are currently 10 Ballers in the community as assigned by their tag on Discord.

Ballers are currently elected by the team at Balancer Labs, however, there intends to be a formal governance process by BAL holders surrounding the election of future Ballers and the encompassing committee.

With that, The Ballers group is expected to expand and rotate over time, encouraging regular community members to keep a high level of engagement so they can eventually become Ballers too!

BAL governance has the ultimate power to elect and dismiss Ballers through Snapshot votes. Ballers and the broader community will constantly assess how participative Ballers are and how good their decisions on the allocation of liquidity mining slots are.

Drawbacks

  • The Ballers Committee is a small group of individuals and can poses centralization risks, collusion, etc. It’s important to note that BAL governance always retains the right to remove Ballers from their position. In addition, becoming a Baller is a selective process and requires a significant amount dedication to the protocol–meaning they’re extremely aligned with the success of Balancer!
  • This structure would eliminate liquidity incentives for the long-tail of pools. This is true, however, the upgrade to Balancer V2 drastically increases gas efficiency and creates a natural incentive for project teams to drive liquidity to Balancer given its competitive conditions for trading, namely lower gas costs. Importantly, some competitors don’t offer liquidity incentives at all and still have sufficient liquidity for the long-tail of assets. We believe that V2’s upgrade is substantial enough to compete with this market dynamic.

Summary

Balancer V2 liquidity mining will operate in 3 tiers. Tier 1 will be voted on by governance on a recurring basis while Tier 2 and Tier 3 pools will be strategically assigned by the Ballers Committee based on community sentiment.

It’s important to highlight that certain pools may be allocated more than one slot by the committee. For instance, the BAL/ETH pool may be elected multiple slots in order for the APY to tentatively match the BAL/ETH APY on V1 as mentioned.

All community members will always have the ability to signal their support for liquidity mining inclusion for T2 and T3 pools in order to guide the committee.

The transition from V1 to V2 will take place over a targeted ~8 weeks with BAL liquidity mining allocations incrementally migrated to V2 on a weekly basis. Throughout the entirety of the migration, the weekly BAL distribution will remain constant at 145,000 BAL per week spread across both Balancer V1 and V2.

8 Likes

Really nice work addressing all the above concerns.

It’s tough to squash every nuance, but this feels like a really nice foundation for us to build off of us and iterate on.

Love the transition to T1 being voted through BAL governance and T2 and T3 being slotted by Ballers.

The target BAL APY feels great and leaves a lot of flexibility to adjust through additional slots in the future.

Keen to see what other community members pop up to join the Ballers. More context on this group here for anyone wondering more about this role and how the name came about!

Overall, I’m in favor of this proposal and will be voting in favor.

1 Like

my only concern is the APY targeting for bal/eth. maybe my numbers are wrong but I estimate ~68k BAL weekly is required for current 70% APY. That would require BAL/ETH to have 10 Tier 2 allocations + 1 Tier 1 allocation (5k BAL x 10 + 15k BAL = 65k BAL).

is this a realistic expectation? may be setting ourselves up for problems later by saying we’re committing to target this kind of APY.

@DavisRamsey is correct.

To maintain the current APY, the BAL/ETH would require around 60.5K BAL tokens per week. This is of course subject to changes depending on pool size and BAL price.

Wouldn’t be easier to simply allocate a fixed BAL supply to the native token pool (in this case the 80/20) without messing around with tiers?

I think this would possibly simplify reward allocations and create a simpler structure.

1 Like

Agreed!

From my perspective, this is just the starting point as we wanted to address the concerns around the significant drop in APYs under this program.

If the community is ok with lowering the target APY in the future (via discussions & on-chain vote), then there’s nothing stopping this system from doing that.

1 Like

it’s tough because our old system design around BAL vs v2 system are wildly different. keeping same APY is fitting a square peg into a round hole lol. I just worry committing to 60-70% kind of cripples our options but as you say can always be fixed later with another vote. That will be a fun one - vote to earn less money on your voting tokens :slight_smile:

1 Like

I honestly think taking more than a few incentivised pool slots for the 80/20 bal eth pool is a bad look and not healthy for the growth of the protocol. We need to be attracting a community that wants to be a baller for the long term, not because of a 70% APY. If a baller is not willing to stick around for a 30% APY, I start to question whether there is a true belief in the Project.

1 Like

Seeing some mixed reviews here.

That’s exactly right. Trying to map it over 1:1 is basically impossible so this is our suggestion at the best way to keep roughly the same flow with new elements at play.

We’re not committing to 60-70% APY, we’re targeting this as a metric to rally around. There is no way to peg it to a specific APY but this is meant to show the BAL community that our intent it to keep BAL as the highest rewarded pool of the lot. It’s not about reducing BAL APY, it’s about gradually distributing those rewards to other tiers and pools as it makes sense.

Want to be super conscious of the timeline we’re on here and want to make sure we vote on something ahead of V2 launch.

This new version was made to directly address the request for a higher BAL APY. While I agree with the sentiment here it’s a valid point that LPs could pull liquidity if they see BAL APY decline drastically. I believe this plan is the best first stab at a new program to be iterated on as we migrate from V1 to V2.

I worry that if we keep fielding comments in this fashion it’ll be tough to get to something we all agree on.

I lobby to bring the revised proposal to Snapshot and let the BAL do the talking!

2 Likes

I agree @Coopahtroopa. We will never find a proposal that everyone agrees with. I think the community did a great job at exposing the different trade-offs and I feel the consolidated proposal is a good first stab at this. As mentioned, we can always vote to change it if there is something sub-optimal.

Regarding @bakamoto20’s excellent comments, I also agree that the long tail of pools/tokens is very important to Balancer, but we can get to incentivize a lot of them by rotating the many slots we have for tiers 2 and 3 (24 in total). And over time with our important efficiency gains in V2 hopefully projects will see Balancer a healthy alternative to Uniswap (our trades will very likely be cheaper than both Uni V2 and V3).

4 Likes

Thanks everyone for weighing in!

Snapshot proposal is up - let’s see what happens :slight_smile:
https://snapshot.org/#/balancer/proposal/QmeCrSe5xL5YWB17TjxAofYi3QnBMyHCYqufD8t3Zonj4S

I think the new proposal does a good job of addressing many of the concerns raised by the community, and I also support the idea of starting somewhere and iterating over time to an ideal solution. It’s never easy herding cats, so kudos on keeping things moving forward!

As we move forward, it may be helpful to write down some guidelines that explain how liquidity mining serves the long term goals of the balancer vision. As the visionary leader of this project, I think you are in best position to help guide the Ballers on what success looks like, and how it ties to the white paper (eg use LM rewards to incentive strategic partnership with other platforms in the ecosystem to tie Balancer into a broader DeFi community)

It would be helpful to have a clearer picture of who we believe Balancer will primarily appeal to as a platform: Is it retail investors wanting to hold a diversified basket of tokens? Institutional Capital? Startup projects and communities looking to generate liquidity via bootstrapping? Strategic integrations with ecosystem partners (eg Aave security module)?

What are the true measures of success? Number of tokens supported by the platform? TVL? Fees generated? Number of LBPs per year?

It’d be interesting to hear thoughts on the above, and see how this evolves in the short, mid and long term.

I will be voting “yes” on the above and retain my full position on Balancer. The thoughtfulness of the community leaders continues to impress. Good job to all.

3 Likes

Are there currently plans to include the Aave/Eth smart pool in the liquidity mining?

Thanks so much for your message @stolos1 and my apologies for the long delay for my answer!

The main objective of liquidity mining IMO is to get BAL in the hands of the right stakeholders. The definition of “right” here is loose on purpose, it’s not up to me or anyone individually to decide what the right thing for Balancer Protocol is. This has to be achieved through long discussions and constructive argumentation like what happened in this thread. Of course a very positive by-product of liquidity mining is liquidity that allows traders to have good conditions on Balancer which in turn generates fees to LPs. It’s a very positive flywheel effect.

Let me try to answer each of your points one by one:

As we move forward, it may be helpful to write down some guidelines that explain how liquidity mining serves the long term goals of the balancer vision. As the visionary leader of this project, I think you are in best position to help guide the Ballers on what success looks like, and how it ties to the white paper (eg use LM rewards to incentive strategic partnership with other platforms in the ecosystem to tie Balancer into a broader DeFi community)

I think you summarized it really well! The objective/mission of Balancer Protocol is to become the primary source of liquidity in DeFi. There are different ways to achieve that: do we want to focus on the most important few trading pairs like ETH<>DAI or do we want to empower the long tail of pairs for new projects and protocols that need liquidity. I think we need a bit of both. First and foremost we want to make DeFi more inclusive and make sure that anyone can participate in a permissionless and censorship-less manner.

So we do have to care about the long tail, it’s part of our DNA in my view. However for the long tail tokens to be well connected with the main assets (WETH, WBTC, DAI, USDC etc) we also want important pairs to be very liquid on Balancer (like WETH<>DAI). This allows for a project TOKEN that is paired with WETH to instantly get liquidity against all the other major tokens for a very small additional gas cost per trade, thanks to the efficient multihops that are enabled by the V2 vault architecture.

It would be helpful to have a clearer picture of who we believe Balancer will primarily appeal to as a platform: Is it retail investors wanting to hold a diversified basket of tokens? Institutional Capital? Startup projects and communities looking to generate liquidity via bootstrapping? Strategic integrations with ecosystem partners (eg Aave security module)?

I’d say it’s all of the above and much more! We have found a very clear product market fit with our LBPs and also through our flexibility allowing other major protocols to use Balancer as a building block in their own projects (like Aave, BZX, Catnip, Ocean, Element, PowerPool and many more). Institutional capital will definitely be a major topic as more of TradFi hears about and migrates to DeFi in the next 2-5 years. We thought of that when designing Balancer V2: the flexible implementation of pools allows for regulated institutional liquidity to make sure it’s possible for them to abide by their local jurisdiction law: they can have AML and KYC in place and an allowlist to ensure only regulated trades/liquidity provision takes place.

IMO there is no way we can redefine the financial world if we do not open the discussion with regulators and the current tradFi world to make sure regulations will be respected.

What are the true measures of success? Number of tokens supported by the platform? TVL? Fees generated? Number of LBPs per year?

All of these are good measures of success! I’d say mostly TVL if we think of our mission “to become the primary source of liquidity in DeFi”. This can only happen if the platform is sustainable and that means LPs are being duly rewarded in swap fees for their risk (impermanent loss). In my view a very important next step for LBPs will be to retain the liquidity of these projects after they finalized their LBP. They should stay on Balancer for the long term, not just for the initial token sale. Ballers will consider this and probably work with projects to make sure that they have an extra incentive (a Tier 3 slot in V2 LM for example) if they agree to stay for a certain period on Balancer V2 as opposed to going to another AMM after the LBP is over.

Again thanks so much for your participation (and from everyone else in this and other threads). This IMO is what makes Balancer great and will be the main factor for our long-term success: smart/engaged community members who foster constructive and strategic discussions like this!

3 Likes

Yes, we will keep LM for the safety module pool using the current V1 while we work with the AAVE team to migrate the AAVE safety module pool to Balancer V2. This won’t be done very soon as it involves a lot of moving pieces.

Thanks so much for the reply. Your long term vision and your pragmatic approach to the big picture with an eye to the future makes tons of sense, and gives me confidence that Balancer will be a market leader for many years to come. Glad to be part of the movement :slight_smile:

3 Likes