[AIP-42] Maximizing Value of Aura LP Incentives & Bootstrapping L2s

Proposals
1

Authors: Lamentations, 0xMaha, Fry

Summary

This proposal makes reward distribution to L2s more efficient, and brings AURA LP incentives in house, by (i) near immediately distributing AURA pro-rata to the gauges voted on on the bi-weekly snapshot votes, (ii) reducing the baseline number of AURA that is minted per BAL for all LPs, and (iii) favouring distributions of AURA to more-aligned parties more likely to participate in Aura governance.

Background

Aura currently emits AURA rewards indiscriminately towards all Balancer LPs that use the Aura protocol. In exchange, Aura DAO directs a share of the BAL earned by these LPs toward Aura stakeholders. This has allowed Aura to represent a large share of Balancer mainnet TVL, along with over 60% of all BAL emissions.

The current system allows Balancer stakeholders that are not aligned with Aura DAO to acquire large amounts of AURA, without providing direct value to the protocol. An opportunity now exists to optimize AURA distribution in order to entice users to increase vlAURA accumulation and spend on voting incentives, increasing participation in the Aura DAO and the governance utility of AURA itself.

Even with the current fee take, LPs on Aura earn 1.5-to-2 times the amount of incentives LPs on Balancer earn.

Aura DAO can vote to reduce the AURA minted per BAL earned for all pools, resulting in a significant decrease in new AURA float. To compensate for this reduction in emissions, Aura DAO can leverage the AURA allocated to the treasury and direct it to pools pro-rata based on their share of vlAURA gauge weight each voting cycle.

In tandem, these two changes will have the effect of:

  • Bootstrapping L2 gauges faster - Currently, there is a 2 to 3 week lag between a gauge going live and getting vote weight on L2s and reaching a steady-state of expected yield. With this change, AURA is allocated to the gauges near immediately and helps to bootstrap the BAL.
  • Decrease the overall AURA emission - allowing for less dilution of vlAURA votes
  • Increasing voting incentive capacity for vlAURA, enabling further onboarding of pools of prominent assets from protocols like Stargate, Synapse, Rocketpool, Swell, Raft, Morpho, etc. This has become increasingly important as Aura has begun its push onto L2s, resulting in large amounts of demand for voting incentives but not enough capacity
  • Utilizing the currently untapped AURA treasury budget, while maintaining budget for other potential expenses like grants or bootstrapping other Aura Service Providers
  • Distributing governance power to “better aligned” LPs

This model is predicated on the assumption that Aura will continue to boost a large share of Balancer liquidity into the future. In our opinion, this assumption is fair for the following reasons:

  • i) under this proposal yield on Aura will remain strategically higher than yield on Balancer directly, even after any fees to Aura stakeholders
  • ii) Aura’s L2 launch should allow it to continue to grow its share of BAL earned, as prior to the L2 push, Aura was unable to capture emissions directed to other-chain gauges
  • iii) Aura continues to maintain dominance in business development / partnership discussions such that TVL, incentives, and co-marketing efforts are focused on Aura vs. Balancer directly

Terms

The two levers—a reduction multiplier on AURA minted and use of AURA to ‘boost’ vlAURA-aligned gauges—are flexible on what governance decides. Based on our analysis of the current yield landscape and factoring in Aura DAO’s needs, we believe this setup makes the most sense:

  • Set the RewardMultiplier on all pools to 0.4, meaning a 60.0% reduction in baseline AURA minted
  • Allocate 180,000 AURA from the DAO treasury every voting cycle to be distributed pro-rata to pool LPs based on their vlAURA gauge weight
  • Reduce AURA/ETH voting incentives, as per AIP-26, from 40,000 AURA in incentives per epoch to 30,000 to factor in increased yield from the above two changes. Keep extra tokens in treasury
  • This proposal should be revisited and re-ratified 6 months after implementation to assess spend on treasury and the market landscape

Assuming this passes in the forthcoming voting cycle, these changes would come into effect for the next vlAURA gauge voting round between Aug 17-21, with multipliers being changed on the 23th and AURA distributions from the treasury from the 24th onward. This is in line with the existing Balancer epochs.

For clarity: the AURA minted against BAL on the auraBAL single-sided staking pool will not change.

Outcomes

This should resulting in the following changes per our estimations, assuming existing system parameters such as % of BAL earned by Aura:

  • An increase in voting incentive capacity by ~15%, or approximately $60,000 worth of voting incentives per cycle
  • A decrease of net AURA supply being distributed by ~33,000 a week or 1,750,000 AURA a year
  • Effectively capping AURA minted against BAL much lower than 50,000,000 units, as specified in the initial tokenomics.

Illustratively, this would have the following effect on pool yield as of numbers gathered in early July:

  • ‘Aligned’ rETH/ETH pool: 4.7% APR (current, once yield/TVL ramps back up) to 5.28%
  • ‘Unaligned’ BADGER/rETH: 26.4% APR to 19.21%

It is important to remember that under this proposed program, all LPs—even those in pools without vlAURA gauge weight—would still yield more than they would directly staking on Balancer.

Regarding use of treasury funds, the treasury will still have a large amount of AURA balance, especially factoring in the 1.5m AURA of protocol-owned liquidity in the 80/20 AURA/ETH Balancer pool that could be recovered by moving to a 50/50 pool.

Implementation

  • Code
  • An Aura incentives multisig will be transferred 180,000 AURA per voting cycle (in batches to reduce overhead), that will then be allocated to the newly deployed GaugeVoteRewards contract which handles pro rata distribution to gauges for each epoch
  • The Aura Protocol DAO will:
    • Call setRewardMultiplier on the Booster for all pools to update the reward multiplier from 1.0 to 0.40
    • Call setVoteDelegate on the BoosterOwnerSecondary contract and set the voteDelegate to the newly deployed GaugeVoteRewards contract that will handle the pro rata AURA distributions by wrapping the call to vote_for_gauge_weights
  • The setup process for the GaugeVoteRewards is as follows:
    • Set the reward per epoch to 180k
    • Call setPoolIds with a start value of 0 and end value of Booster.poolLength()
    • Add veBAL, veLIT and veUSH as no deposit gauges
    • Set dstChainId to 110 for arbitrum gauges
    • Set dstChainId to 111 for optimism gauges
    • Set dstChainId to 109 for polygon gauges
    • Set the childVoteReward contracts for each chain
  • To ensure that earned rewards are not effected by the reward multiplier reduction A script will run to claim existing rewards on behalf of all users with 5 BAL or more of rewards to claim. It will loop through all holders on every reward pool and call BaseRewardPool.getRewards(holder, true)
  • The parameters will be reviewed 6 months following the first gauge vote
11 Likes
2

I highly support this proposal for a few reasons
-Increasing the gauge voting market by almost 20% is a net positive to all Aura lockers.
-Reducing the inflation of Aura by 27,750 Weekly AURA is highly positive for all current holders and lockers.
-Lowering the total possible supply of aura against Aura creates a lower fully emitted supply.
-An increase in yield would give depositors more reasons to migrate to Aura, thus ensuring a higher veBAL capture rate.

3 Likes
3

This is a very interesting and well thought out proposal. One question comes to mind though.

Has anyone thought about what this would do to the yields on auraBAL? It seems that under this proposal less AURA in total per week would be emitted to LPs, which would result in some expected reduction in auraBAL yields. Any idea how much?

Is that going to make auraBAL less attractive, it’s been growing nicely as of late? Is there a way to compensate?

2 Likes
4

Interesting question that we should play out a bit.

So if this passes, we are now going to have Aura-aligned pools (“aligned pools”) and non-Aura-aligned pools “unaligned pools”). Aligned pools receive additional treasury AURA. Unaligned pools don’t.

My first thought concerning your inquiry: the stated assumption that there will be some expected reduction in auraBAL yields isn’t necessarily accurate, though it is possible. The auraBAL pool will still mint AURA at the previous rate, so if Aura earns the same amount of BAL, then the auraBAL pool will have the same APR.

This is because auraBAL and vlAURA yields are supported by 24.5% of the BAL that Aura collects.

So, the fact that less Aura will be emitted to all BPTs staked on Aura doesn’t necessarily mean we will see a decrease in Aura’s BAL intake.

There are quite a few different factors to consider.

Before jumping in, let’s reinforce the concept that if this proposal passes, even for unaligned pools, Aura will still provide more aggregate emissions for BPTs staked on Aura than if the same BPT was staked only on Balancer.

Now let’s play with a non-exhaustive list of factors that come into play here:

  1. Will there be a net decrease of BPTs staked on Aura for unaligned pools if Aura is still providing more aggregate emissions to them than the same BPTs would receive if staked on Balancer alone?
    Maybe yes, maybe no. I’m inclined to think probably not, but it’s likely worth monitoring.

  2. If we assume the answer to # 1 is yes, does that also mean that fewer BAL emissions will be collected by Aura overall?
    Not necessarily. For a few different reasons outlined below, we could see AURA’s BAL intake remain relatively the same or increase. If the answer winds up being no, then the stated concern doesn’t seem like it will come to fruition

  3. We may also see increased BPT TVL for aligned pools. If we do see an increase, it’s possible that increase Aura’s BAL intake. It’s also possible that the BAL intake would remain relatively the same. In either event, the stated concern doesn’t seem like it would come to fruition.

  4. A - Let’s assume the answer to #1 is yes and there is a decrease in Aura’s BAL intake for unaligned pools.
    B - Let’s also assume the answer to # 3 is yes and there is an increase in Aura’s BAL intake for aligned pools.
    The next question is: does the increased BAL intake on account of 4.B outweigh the decreased BAL intake on account of 4A?
    If the answer is yes or there is no overall net change, then the stated concern doesn’t seem like it would come to fruition.

  5. What will be the consequences of Aura’s expansion to L2s coupled with the implementation of crosschain veBAL boost in relation to Aura’s aggregate BAL intake?
    If as I suspect the BAL intake increases because of the L2 expansion, and
    C - it increases in a manner that offsets or surpasses any decrease associated with assumption 4A, or
    D - it, coupled with an increase in BAL intake under assumption 4B, offsets or surpasses any decrease associated with assumption 4A, then the stated concern doesn’t seem like it would come to fruition.

2 Likes
5

Sounds promising! Glad to hear we’re working on this problem.

5 Likes
6

Hmm ok, maybe I’m missing something.

Given that:

  1. The AURA that flows to auraBAL comes from a 20.5% fee that is charged to pool depositors and paid to auraBAL stakers. The docs actually define this as 20.5% of BAL, and then I assume AURA is minted on top based on the formula that is changing. Fees - Aura Finance.

  2. The plan as it states alters the formula that the docs seem to imply is used for paying AURA alongside the BAL fees collected from LPs and paid to auraBAL holders.

  3. A stated expected decrease in total AURA emissions of all AURA distributed.

Maybe 1&2 are just about updating the docs to explain that 20.5% of all AURA distributed to LPs regardless of the reason/source flow to auraBAL. but 3… I really am having a hard time understanding how the auraBAL yields don’t go down.

I guess your point is that once this happens there will be lots more LPs using AURA/more of the BPTs will be staked on AURA which should result in everyone getting paid out more, which will make up for the difference. Somehow I don’t get that sense. Does this mean that that 1.5 million AURA per year will actually be minted due to aura having a higher capture Balancers staked LP AUM?

This plan is good, and I don’t think the intention is to pay less AURA to auraBAL, but as stated it seems like that will happen and it would be good to add something to ensure it doesn’t, by paying some of the aura saved out to auraBAL if yields drop. Maybe there is some other math that was planned/part of models/simulations that is not clear in the proposal.

Has anyone modelled the situation/change out at all? We seem to be coming to very different conclusions from the same math. I just don’t understand how a net reduction of 1.5 million AURA per year and no increase in fees is going to result in the seem amount of AURA being paid out to auraBAL, and I don’t see any flows proposed to fill that gap.

1.44 million AURA is to be saved by the program in the first year. Would paying say 15% of that to auraBAL make up for most of the impact, while still trying to save a bit on auraBAL.

So that would be 108k AURA over the proposed first 6 months?

1 Like
7

Im confused by the part of your question that keeps referring to AURA emissions as being subject to collection and redistribution to auraBAL.

Is it not 25% of BAL intake that is collected and redistributed to auraBAL, vlAURA, and for harvesting?

1 Like
8

Cool idea. Enjoying reading the perspectives but conceptionally this seems like a good path forward if aurabal yields really will not be dramatically altered

2 Likes
9

image
image1438×1982 206 KB

So I always assumed that the BAL being paid via auraBAL also caused minting of AURA along side it as per the formula. Maybe I misunderstand, it’s not super clear.

The point remains: Half the auraBAL yields are in AURA. I’m pretty sure some how or another they are getting a 20.5 of all AURA minted and emitted via the current gauge system. In my opinion, it is important that this remains the case, and that if there really is to be a lot less AURA emitted, some extra AURA is paid to auraBAL lockers to keep yields high and auraBAL growing.

2 Likes
10

Sounds like we should just reduce the Aura paid to LP stakers but not to AuraBal stakers.

4 Likes
11

@Tritium - This was my understanding. Now clarified in the Temp Check.

3 Likes
12

An excellent proposal that:

  • Gives extra incentives to be an Aura-aligned LP
  • Keeps AuraBAL attractive
  • Enables the increase of voting incentives for vlAURA holders
  • Reduces Aura inflation

I strongly support it, as it makes perfect sense and covers pretty much all the needed bases.

3 Likes
13

Extremely interesting and thoughtful proposal. Could we have some more details on what would qualify a pool as aligned? Ideally the criteria should be fetchable on-chain so we can avoid bias.

3 Likes
14

The ‘aligned’ pool term was just referencing pools whose gauges have vlAURA votes

4 Likes
15

I believe moving from a 80/20 to 50/50 pool for Aura/ETH is a very interesting move considering the current needs for liquidity on both sides of this market, it could also participate to trigger a reversal price trend.

Moreover the proposed changes in Aura emission make total sense to enhance the flywheel with vlAura voting utility.

The future is great for Aura multi-chain !

4 Likes
16

https://vote.aura.finance/#/proposal/0x099d75d2da1bdbb38aaa9fd86c1f63cb589d165c2b31387945e8acb598a281f5

3 Likes