Long-term Liquidity Mining Rewards solution by adopting a modified Ve model.
B3 is launching soon, and Bancor will be going multi-chain shortly after. In order to maintain deflationary pressure on the BNT token, it’s imperative that we pursue a long-term LM solution that strategically limits inflation while providing additional utility and strengthening community engagement.
Problem: The currently liquidity mining rewards emissions at Bancor can be better aligned with protocol success. The current system could be easily gamed by a 3rd party protocol, by buying/renting vBNT influence. This isn’t unique to Bancor, but we should have mechanisms in place to benefit from this scenario possibility.
Solution: Adopt a modified Ve model, similar to curve so that BNT holders/stakers could benefit from LM emission competition while driving long-term protocol staking.
General Idea: Allow vBNT to be converted to VevBNT by locking their vBNT…stake/locked vBNT becomes VevBNT. VevBNT will have voting power to vote weekly for LM rewards allocation in Bancor. The VevBNT stake/lock periods could operate similarly to the way Convex handles CVX. If you’re familiar with curve, then everything above is very similar. Here’s where I propose differentiating from curve.
**Tie the VevBNT model into the vortex mechanism. Split the vortex burn by 50%. When the vortex burn is triggered, burn 50% of the vBNT and allocate the remaining 50% to the VevBNT stakers/lockers. The ability to earn additional vBNT rewards would be significant motivation to participate in the Ve Model, and drive new demand for long-term locking/staking.
Tie the weekly LM emissions to the Vortex Burn. This could be done in a lot of ways. One idea would be to curve the emission over time so that, eventually the total weekly LM emission would be less than the Vortex burn rate. This would ensure long-term deflationary properties, while guaranteeing that LM rewards could continue indefinitely, by tying emissions directly to the success of the protocol (vortex burn rate).
The solution above also brings back the potential vBNT arbitrage that will no longer be possible in Bancor Version 3. With the new vBNT tokenomics, I’d argue that an arbitrage would be welcomed and even more probable (and therefor useful) in V3. The new type of arbitrage above would exist by creating a vBNT/VevBNT pool. Since, vBNT can be converted to VevBNT at a ratio of 1:1, should the price of VevBNT ever rise above vBNT, arbitrage would be possible. Another arbitrage-like opportunity could occur if staked VevBNT could be traded to vBNT in the vBNT/VevBNT pool. In this scenario you have to factor in the time lock component mentioned in the “General Idea” section above. Once the VevBNT has surpassed it’s time lock period, it could also be converted back to vBNT at a 1:1 ratio. This gives patient Bancorians an opportunity to take advantage of the price difference between vBNT/VevBNT by waiting out the lock period.
This provides a strategic long-term LM solution that aligns with Bancor’s success. This will also incentivize products to be built on top of Bancor. In my view, this is a great way to drive long-term value to the BNT token.
A few implications:
New token projects that want to offer dual side LM rewards in Bancor Version 3 will now have to purchase VevBNT to vote for BNT LM allocation. Better Bancor/Token Team strategic alignment, this will help to discourage dumping farmed BNT.
More equitable distribution of LM rewards. Smaller pools will now be better positioned to get *some LM rewards.
Is there a need to continue LM rewards indefinitely? Rewards were intended to attract liquidity to v2.1, which it accomplished. Investors came for the LM, but stayed for the IL protection. LM rewards served their purpose, and should be retired in favor of other marketing initiatives, such as going multi-chain, which would allow us to grow trading volume, increase liquidity and attract new users (and their liquidity).
I partially agree with you, because we don’t need, and likely will not be able to afford the kind of LM rewards we’re emitting now. However, my proposal aims to have the amount of LM emitted based on a value lower than the Vortex burn. I didn’t detail this very well. In this way, the LM rewards can still exist but will be deflationary because the amount printed will always be lower than the amount burned. Also if Bancor’s volume (and likely TVL increases)…this will increase the Votex Burn and allow for more LM emissions. In any event, total LM emissions would be a TINY fraction of what is being minted currently.
Ultimately people will follow the returns. If LM rewards ended on Link, ETH, WBTC, and the stables, we’d see a massive TVL exit. The Bancor team knew this, and that was why they were so active to prevent Simp’s proposal/vote to ended LM rewards on many of the pools last year. Their position was that ending the rewards prior to V3 was too risky. I tend to think this is accurate while believing that we’ve been severely overpaying for stable TVL. Maintaining a competitive APR on major pools will almost certainly require LM. For example, if ETH APR drops to AAVE levels…People will just go the LIDO/CURVE/CONVEX route. In the end, I believe this strengthens the utility of BNT and vBNT, and will drive price appreciation through voter escrow.
I agree about equitable LM rewards across smaller pools being a shortcoming of V2.1 and possibly V3. LM rewards are also distributed generously to BNT, which I understand helps to maintain market interest in BNT supporting its use as the numiere. I have never quite got my head around why a market pricing mechanism for BNT is so central to the protocol.
Agreed. Our current system is slightly similar to voter escrow. However, smaller vBNT holders/pool are marginalized & disincentivized because their participation doesn’t meet threshold requirements. True Voter Escrow solves this.
Overall, Bancor’s success is pinned to its ability to pay IL from protocol fees. Because of this, BNT’s price is directly related to token emissions (lower price more emissions for IL). In turn, this is related to user staked BNT and GAS fees. Once Bancor goes multi-chain. It’s going to be much easier/cheaper to get in an out of BNT. So, you can imagine how a major IL event on Bancor could lead waterfall event where huge amounts of emissions and selling could really drive negative price action. For this reason, price appreciation is critical to the design.
I am not against the idea of creating a gauge, a predetermined BNT budget that is divided among current whitelisted projects depending on the weight of votes. It would help us budget our LM and spread it out more evenly accross the board. A mechanic like this could be lucrative to encourage protocols to aquire BNT and could be rolled out with a small budget which we could slowly ramp up if we choose to.
However, I think the time-lock and other extra veTokenomics seems unnecessary, the time lock seems unnecessary since we have a regularly super high staked supply percentage, BNT is already locked up for the most part through no forced action because our community is extremely bullish on our token. Furthermore veBNT will just become inconvenient and people will probably gravitate towards a platform like Convex which will just build a wrapper of the token and could then create liquid staking which could lead to a platform like convex basically just owning the entire platform like they now kind of do with CRV (Convex owns something like over 50% of the CRV Supply in their cvxCRV wrapper).
Since there is no need to have a timelock it also means there is no need to have a tiered emmisions system (Scaling APY). Again though would be super in suport of creating a gauge, would also help some of the smaller protocols attempt to get some LM through bribes.
I agree that tiered emission doesn’t make since for Bancor…especially since BNT is elastic. This is why I propose tying emissions to the Vortex Burn. In this way, we can ensure that less BNT is minted than is being burned.
For the veTokenomics, I think it will actually benefit Bancor. You are assuming that BNT lock supply will always be incredibly high like it is today. However, we need to realize that APY on staked BNT is slowly decreasing in the LM Reward pools. What will BNT APY look like in a year, or after B3? Now, we have Bancor going multi-chain, where gas fees will not be an issue. When you consider these 2 factors, you can imagine that farming/dumping BNT would be likely to increase, and even if it doesn’t the, veTokenomics is another mechanism to keep BNT in the protocol. We don’t need an aggressive 4 year curve lock.
I agree, this could lead to a 3rd party application gaining Bancor influence, but I don’t think this is a bad thing. I think it’s inevitable for protocols to build atop other protocols. Bancor needs to be prepared for this eventual outcome. At least the ve system allows participant choice, increases transparency and encourages user participation. In a way, the influence is already possible today. Convex (or another protocol) could create a cvxBNT token today, autocompound…and vote LM rewards through regular governance methods.
This is true, but there would be no point in doing that. they wouldn’t offer a better deal by wrapping it, if anything it would be a lesser form of BNT with more slippage. By adding the time lock you create an incentive for people to gravitate away from bancor and towards liquid staking derrivative protocols.
Even less useful with Bancor V3. Xtoken tried this, and if it weren’t for the hack…, the product was fine…the auto-compounding was good. My overall thoughts about the current LM system…is that it isn’t equitable, and users aren’t getting enough value prop out of their vBNT. We’ve been incredibly fortunate to have BNT staked at such a high rate…and I feel gas cost and contract complexity has really helped in this regard. So, there doesn’t appear to be any consequences for continuing LM on the main pools… but I think this will change, so we need a new responsible emission system.
In my view the derivative wouldn’t be BNT based…it’s vBNT based (veVNBT)…so BNT stays in the protocol, auto-compounding. I think a healthy option would make sure there’s an arbitrage mechanism between the asset and the derivative. The underlying pool between this could be on Bancor or not. A simple arbitrage opportunity would be a short(ish) time-lock on the derivative. This could be fined-tuned.