[Pre-RFC Discussion] LP & BAL holder incentive alignment

Governance General Proposal
1

This thread as an opportunity to discuss some things a few of us have been thinking about with regards to LPs and better aligning incentives between LPs and BAL holders, in advance of a formal proposal being put forward if we gain sufficient traction for a specific approach.

In growing Balancer we want to incentivise as many LPs as possible to become actively involved in the community and marketing the protocol to the wider crypto community, in order to build the Balancer brand and achieve its goal of becoming the top source of defi liquidity.

There are various ideas that have been raised to encourage LPs to further grow the protocol, and hold BAL tokens to incentivise brand loyalty & growing the protocol:

A) Locking mechanisms / APY boosts for locked BAL

We could take an approach that has similarities with Curve, which has worked very well for them. In order to earn the maximum BAL rewards for depositing in Balancer pools, you have to lock up some amount of BAL for a period of time up to 4 years. The longer an LP locks their tokens, the greater their BAL reward boost factor (up to a limit).

The nice thing about Curve’s mechanism is that it also means any yield aggregators that may choose to build Balancer vaults have to lock up some BAL for years to maximise APYs. Their mechanism can be explained here: Boosting your CRV Rewards - Curve Finance

Worth thinking about if we went down this type of path is whether locked BAL also got proportionally more voting power than unlocked BAL (the longer you’re locked, the more voting power your BAL has). This helps to further align LPs & governance with the long term interests of the protocol.

B) UMA’s KPI token utilisation for a share of BAL rewards

Instead of distributing only BAL to LPs each week, we distribute some mixture of UMA KPI tokens (that expire some months into the future, where the greater Balancer V2’s TVL/trading volume is, the more the KPI token is worth, up to a certain ceiling) and BAL.

Trading volume could potentially be more interesting than TVL (encourages LPs to think about how they can grow volume: create/deposit in pools likely to increase volume more, promote to traders), although it would take thought to prevent it being gamed. For example, we’d probably have to ensure only Trading Volume done at a minimum fee level counted, so people didn’t just create ~0.01% pools or wash trade.

I suspect we could design it in such that wash trading would be unprofitable for the minority that would do it, a majority of LPs wouldn’t go to the effort and a lone actor would be wash trading for the entire protocol’s volume just to get gains on their own LP rewards.

More on how this works can be read here: UMA KPI Options and Airdrop. UMA will be airdropping a new… | by Hart Lambur | UMA Project | Medium

I think we could potentially make something like this more successful than UMA did with their initial program, as we’d only be distributing to LPs, the $ amounts would be much more significant to people and our products are easier to understand / much more widely used in DeFi to date (“usage” is a simple LP deposit or trade).

C) Simple vesting of some share of LP rewards

This was discussed in the past. Some share of BAL rewards earned each week can’t be claimed for a period of time, e.g. 6 months.

I’m sure the community has other ideas for what we could do here, too. Of course the trade-off with any of these approaches is that we make the liquidity mining a varying degree of more complex for LPs: it’s a bit more to understand, and we’d have to educate users. Locking also ties people to using particular wallet addresses for the locking period.

Personally I think that the losses in simplicity of rewards could be outsized by the gains in community participation & incentive alignment for any of these approaches, and look forward to hearing people’s thoughts on ideas here or other approaches we could try. The UMA option could be something to try initially, as we could do it for a one-off quarter/couple of quarters very easily to see how it worked, vs. the Curve like path which is a greater overall commitment in terms of development & launch effort.

The UMA option is also something which I’m sure they’d be delighted to work with us on some shared marketing for, and would likely pick up attention in defi as something interesting if we moved relatively quickly to try it before another significantly sized DEX.

2 Likes
2

@chandlerdekock @poopster just tagging you guys here as you brought up the UMA Options in another thread and thought you may want to contribute.

2 Likes
3

Did this ever end up happening? we passed the vote but I haven’t heard anything about it.

My feeling is the KPI options will end up going the same way as these call options did. The pool containing those call options had a total of 9 trades. Granted it didn’t cost us anything to do so I guess no real harm done. KPI Options could be cool, I like the idea of aligning everyone towards increasing TVL.

3 Likes
4

Yeah looks like they weren’t too heavily used. I think using them as part of people’s weekly rewards is a very different application though, as you’re distributing them to all LPs and they can’t do anything with them until they expire for some amount of BAL based on what TVL is at that time.

So people will have to use them. You need rolling expirations for them though, and probably to take TVL average over a period of time to prevent whales from just depositing into pools on the last day of the month, and withdrawing after option expiration.

5

it introduces another complexity in ppl getting rewards. You could expect our rewards APY’s to increase (tvl to decrease) because ppl discount the value of the options they’re getting compared to just getting BAL

2 Likes
6

I mean, either you end up effectively decreasing inflation for a period of time because people de-value the options and TVL doesn’t increase as we’d like (so options expire for less BAL), or people will hold them & work harder to try to increase their value (and thus TVL) before expiry.

I don’t really see either as a major loss from running as an experiment. It introduces an interesting dynamic when people are allocating liquidity to pools too, where you might choose Balancer over somewhere else to help boost the value of your existing KPI tokens that are expiring soon.

2 Likes
7

What about just issuing the options to all LP’s on v2 and allocate BAL from ecosystem fund? instead of taking from weekly rewards

8

My concern with this is that the options we could afford to issue with the ecosystem fund won’t be worth enough for the LPs to care, not enough skin the game vs. e.g. a quarter of their rewards which would be pretty significant for many LPs.

1 Like
9

feels like beating LP’s with a stick saying get out there and generate TVL, lol

25% of weekly rewards feels way too high for this to me. 25% would be we’ve confirmed people like this and it works.

1 Like
10

I think it has to be high enough to prove it works though, there’s no guarantee but if you go small it definitely won’t work as no one will care.

11

We have new KPI options in the approval / voting stages now but the metrics are different this time. These are only givin to SuperUMAns “I think” than based off how many integrations with other protocols we all achieve with a cap of course of 30 “I think”. Their are a lot of protocols that want to use the KPI options to build their TVL back up , gather their troops and start climbing that hill back to the top. I started writing UMIP KPI price feeds for protocols because there are so many coming in right now.
UMA has grants issued to Universities around the globe to be studied and tested. Not only Universities but, Behavioral schools and Alternative schooling as well are using KPI Options as incentives for behavior , GPA , attendance , games, ect .

The KPI Options have to be dealt out correctly from the start to those that will hold till the expiry. Wether using air drops or distributed on the fly for tasks or achievements you have to get them in good hands regardless of the metric is that is chosen for them to work as anticipated. They get in the wrong hands you will have users dumping them , than whales trying to buy as much of them as possible than dump them causing the value to decrease. You can look at them like this. If a couple people steer the ship off course everyone suffers. That’s why UMA limited the amount you received and Air dropped the in brackets or tiers, if you will.
120 for minting and pooling
90 for -------holding over 20 UMA
60 ------- Contributing to Discourse/discord/twitter,ect

I was in a call with a protocol and heard of this buying large amounts and dumping them while the price was up , I sensed there was more to the story as well by the way they addressed his absence on the community call.

If wishing to seek other protocols experience on how the first round of KPI Options went, there is a well written article about Badger DAO and their experience. They just decided to go forth with a second round. It is a very informative article including some data and numbers from the first KPI Options. Also, changes they made and why. You can check it out here:
BadgerDAO rebase mining & KPI Options pt.2
Good read. I think KPI options make sense and there are many many ways to use them. Everybody wins. The community is having a blast, everyone knows everyone , there is something for everyone to get involved in and help in their area of knowledge say: social media . So they have their calls scheduled they make during the week between each other. Than there are calls everyone can hop on to ask anything , ask for help on something , product development, ect. Starting next week there is a trivia night and anyone can join.

Balancer is and has been the utility for UMA’s yDollars/ UMAbills.
Earning rewards for your minted UMABILLS (wETH, June) requires you to also deposit to the UMABILLS Balancer pool. Not only does this qualify you for UMA rewards, but it also entitles you to receiving Balancer rewards too! You can easily use this utility to either add liquidity to the pool (by posting UMABILLS and USDC) or remove liquidity you have already added to the pool.
*Yield Dollar (wETH, Jun)

  • Yield Dollar (wETH, Sept)
  • Yield Dollar (wETH, Dec)
  • Yield Dollar (renBTC, Jun)
  • Yield Dollar (renBTC, Sept)
  • Yield Dollar (renBTC, Dec)

I heard of Balancer stories on Redditt. I ended up using Balancer through UMA , minting YD-ETH MAR21. I escaped the familiarity of day trading for years on Coinbase Pro. I was in a brand new world. blindly trying to swap, figure out how to get coins into a pool, not knowing about gas , approving tokens, the proxy , I got destroyed. I found discord forums to be useful and began asking questions. I read others questions and the answers to their questions in this very forum as means of survival in this new expensive world of DeFi. To this day I am in the UMA and Balancer communities. I am a prime example that when two protocols, working side by side and leveraging each others products it provides the sought after exposure to new users and community members. That being said, onboarding is just one more great benefit the KPI Options will bring to the table.
I strongly believe that KPI Options would be of great assistance and will be beneficial to Balancer, from the treasury to the community. I know it works and I’ve seen it work. The 19th of May is behind us ,it is a buyers market out there, we all know that Balancer can do $ 50, $60 and $70 per token. I hope we can at the least try a smaller 30 day KPI Option and come to a conclusion if incentivizing your community for participation actually works.

1 Like
12

Hi folks - First off, by way of introduction, I’m Stefano from LongHash Ventures. We have been working closely with Balancer Labs through Asia DeFi Network on a few topics, including Community Building in Asia, Liquidity Mining, and Governance.

A week or so ago we started drafting a RFC/Draft Proposal on BAL staking, taking inspiration from the staking mechanisms of Curve, Idle, and Pickle. This, although worked on in parallel, builds on the idea (A) outlined above by Bakamoto20.

We think there’s two overarching benefits to BAL staking, as we have articulated, namely:

  1. It rewards BAL holders who take the long-term view of the protocol and choose to lock up their tokens
  2. It is a starting point in the journey towards governance decentralization

We chose to post a separate RFC as it only touches on idea (A) outlined above and we believed it is fleshed out to stand alone on its own merit. We welcome all your inputs, feedback, and hopefully a rich discussion around it so we can improve it and flesh out the nitty gritty details to take it forward.

The link can be found here: [RFC] Staking of BAL for Economic and Governance Benefits

3 Likes
13

Thanks Stefano, great work. Given this, we can probably retire this thread for now. Perhaps UMA KPI options may be relevant within a different future proposal, but I think it’s definitely too much complexity to combine locking with the KPI options for LPs, and the locking proposal was my preference for addressing this alignment challenge anyhow.

3 Likes
14

The vote was too close to the expiry date so we were not able to actually implement it in time. The UMA team (@Clayton_UMA) was planning on creating another proposal that would make it possible to repeat the process every month without having to pass new votes every time (IMO too much governance overhead).

3 Likes
15

I wouldn’t think any of these options are mutually exclusive. Overall the KPI options seem the most interesting and potentially value aligning. That being said, I’d worry about the KPI options being gamed both in the aspect of achieving the KPIs set in the option contracts as well as gamed via pump and dump efforts. Besides the Badger DAO example and UMA themselves are there any other good examples of the KPI options being used in the wild?

1 Like
16

A bit chicken-and-egg, because there wasn’t much trading because there wasn’t much liquidity. Indeed, I dropped the ball on this.

KPI Options are quite a lot different. They energize the community to achieve an outcome – You’re offering them pure upside in exchange for their effort to help an outcome.

Call options are just a risk transfer product. I intend to have the Super Umans (who are UMA’s own KPI Option recipieints) come present a new proposal per @Fernando’s point, as it will count towards their metric of uDAO integrations.

3 Likes
17

I am absolutely not a fan of these restrictions. But if I have to choose between the three options then I would go for Option A.

But if I lock for example 50% of my rewards then this would mean that I am not able to compound my rewards in the Bal 80/20 liquidity pool, so why don’t we do here a different approach:

  • instead of locking the BAL rewards you instead lock 10 to 100% of your provided liquidity in the pool for 3 months or up to 4 years.

or we offer both:

  • The BAL rewards increase if you either lock 100% of your BAL rewards or 100% of your liquidity or something in between (see below)

So for each pool you have this kind of gauge in the liquidity pool UI:

balrewards
balrewards1785Ă—1080 88.5 KB

It lets you customize the lock of liquidity and rewards. So you can decide for example to lock 31% of your liquidity and 69% of your rewards or 0% of your Liquidity and 100% of your rewards.

Optional feature is to auto-compound the rewards. This feature could be for example limited to BAL liquidity pools only

3 Likes
18

Thanks so much for this thoughtful suggestion @Nwar and also for putting the time to build this nice UI mockup, appreciate it!

I think it could be a bit to complicated and we all know that in this space simplicity and easy UX is extremely important. I wonder if we could start with experimenting with one thing at a time and measure its impacts as opposed to changing a lot of things at once and risking not knowing what worked.

Not sure how hard technically it would be to do this, but maybe a good start would be simply to have a boost in BAL liquidity mining for LPs that lock their BPTs in the staking contract (which is being finalized atm by the BLabs’ team and should be live in the next months). Something like:

  • 3 months lock → 10% boost
  • 6 months lock → 25% boost
  • 12 months lock → 60% boost

Just some random numbers to get the discussion going.

PS: this is not an official stance of BLabs, these are just my personal thoughts

4 Likes
19

This is interesting, are you suggesting only doing this for the 80/20 BAL pool? Something to think about with locking the BPT like this is that reward allocations for that pool could fluctuate within the LM program after the lock: so someone locking up for 12 months and a 60% boost could be disappointed if weekly BAL allocation was reduced a lot for the 80/20.

I can see it working best maybe if you just had to lock a certain amount of the BAL 80/20 BPT per $1k liquidity you have in pools or something like that to earn the boost, in a manner similar to how you described. The boost then applies to whatever you are pooling inside the LM program, and we use the LM program to focus people on providing liquidity where it’s needed most.

So e.g. you stake $10k BAL/WETH 80/20 for 12 months, and that gets you a 60% boost on the APYs you earn for any other BPT you own, up to a total of $50k pooled (I don’t know what would make most sense for ratio of locked:other pool reward bosts, that example is 20%). If someone then wants to pool another $50k with max APY, they lock an additional $10k more BAL/WETH 80/20 for 12 months to get max reward, etc.

We’d probably have to fix the 80/20 BAL pool’s weekly BAL in place, although I suppose we already do have a floor with the 15k floor thanks to T1, so people staking would know that no matter what happens in LM, they’ll get at min their share of that over the locking period. It might be that the locking reduces the need for BAL rewards going to the 80/20 pool, as people will want to lock for the benefit of earning more from their other pools.

I have no idea if I’ve explained myself clearly, hopefully that makes sense! It’s late :sweat_smile:

1 Like
20

Balancer is losing the plot.

As a long time (and decently large) LP I can’t help but feel Balancer is failing to live up to its potential and is now starting to flail.

Why is there dev time spent on fracturing liquidity and farming rewards onto Matic when the basic portions of the much vaunted V2 architecture doesn’t exist for end users?

If mult-hop transactions are gas savers and an alluring technical feat, why can’t I take my weekly Bal rewards and in a single transaction re-invest them in the ETH/DAI pool? This seems like a basic of use cases for the V2 single vault system and it’s no where to be found.

Where are the delegators? Why isn’t a portion of the single vault delegated to Yearn vaults? Are tokens in Aave? If not, why not, and if they are why isn’t that more publisized?

Why is there any discussion of bizarre UMA / KPI cross-over airdrops, locking up Bal rewards (as if that’s an incentive), when the APY’s on the Investment Pools are consitently dropping?

I’m sorry but Balancer is sliding backwards and fast.