B.Protocol Governance Token

EDIT 1: Per the suggestion of @amplice and others, the token was renamed from BRO (as initially suggested by @sunny) to BPRO.

EDIT 2: As many members requested to reduce the collateral to debt reward, it will be set to 20% for collateral and 80% for debt.

EDIT 3: A clarification that the Genesis backstop consist of 3 members.

EDIT 4: The community raised the discussion to a vote on snapshot.
Maker users can vote here. Compound users can vote here.

EDIT 5: the mining rate per block was updated to 0.47564688 BPRO per block to reflect 250k BRO minting per 3 months.

==============================================================================

The community decided to tokenize B.Protocol governance, and it seems there is a consensus that the initial Genesis governance will be divided equally among current Maker and Compound users.

A week before the binding on-chain vote for the new governance, now is the time to discuss how the Genesis governance will evolve to include more users, developers and liquidators (via token minting).

Other than the token distribution, the community should also reach a consensus around the decision making process post the Genesis vote, e.g., off-chain voting with multisig execution, and how future Jars will be distributed.

We start by considering a plan for the token distribution, then discuss a mechanism for future decisions, and finally suggest a way to distribute future Jars.

The objective is to reach a consensus and make the needed adjustments before formalizing it with an off-chain voting, and then to put it into an on-chain vote on April 26th.

Token distribution

The name of the token is also subject to the community decision, but for simplicity we denote it by BPRO (EDIT: credit goes to @amplice and @sunny for suggesting this name).

Initial distribution

  • 500,000 BPRO to Current (*) Maker integration users, proportional to their mScore. E.g, a user that has 0.1% of the total mScore will get 500 tokens. (*) snapshot will be taken on April 26th.
  • 500,000 BPRO to Current (*) Compound integration users, proportional to their cScore. E.g, a user that has 0.1% of the total cScore will get 500 tokens. (*) snapshot will be taken on April 26th.

Subsequent Distribution

New BPRO tokens will be minted for a minimum period of at least four years. The minting will occur in every block, and for simplicity we discuss the expected minting according to time periods.

Perpetual minting

  • 1,325,000 BPRO per year will be minted into a Reservoir contract, which will be under the control of the DAO, and will be used to on-board new participants into the governance, in any way the DAO will see fit.

  • 825,000 BPRO per year will be distributed for devs. In the first year this will be sent to an address controlled by by the founding team, but the founding team will not vote.

One time minting

  • 250,000 BPRO for new (and existing) users, for a 3 month period. After that, the DAO may decide whether to use a portion of the Reservoir to allocate governance tokens to new users.
  • 150,000 BPRO for the current Genesis backstop, which is composed from 3 entities, for a period of 1 year . After that, the DAO decide whether to use a portion of the Reservoir to distribute more BPRO to the backstop.

On-chain and off-chain decision making

The governance smart contract will be the one currently used by Compound Finance (Bravo) for on-chain voting. For off-chain votes, snapshot.org will be used, and a selected multisig will execute the decisions voted by the token holders.

On-chain voting

Initially, on-chain voting will only be used to replace the selected multisig, if it misbehaves.

A Quorum of 250,000 BPRO will be set, meaning that in order to approve a proposal, at least 250,000 tokens will have to vote in favor.

A Proposal Threshold of 2,500 BPRO will be required for a user to create a new proposal. Using Delegation will enable users to delegate their voting power to other users, enabling them to propose once the threshold is met.

Once a proposal is placed it will have a two days Voting Delay before voting can begin and a five days Voting Period to reach the needed quorum for it to pass.

A two days Timelock period will be set for successful proposals before they will be executed.

We welcome additional discussion/suggestions on the initial quorum and other parameters.

Off-chain voting

The off-chain voting would be used to give execution instructions to the multisig, in particular on-boarding new liquidators, and on user incentives.

In addition, the multisig will initially execute the DAO’s decisions regarding the Reservoir tokens.

In the absence of an execution mechanism to off-chain votes, it is imperative that proposed decisions will be precise, and to formally define the decision mechanism.

The proposal is that initially only yes/no decisions could be voted, and that a 250,000 BPRO quorum will be needed.

This mechanism could be changed only if a 250,000 PRO quorum will vote, on a yes/no decision, on a new decision making proposal.

Alternatively, a change could be made by voting, on-chain, on a new multisig that will adhere to different rules.

We welcome additional discussion/suggestions on the initial mechanism.

The multisig

It is proposed that the initial multisig will be composed of the 3 genesis backstop members (and to reduce their workload, every time 2 out of the 3 liquidators will serve as key holders in the multisig), and one key will be held by one of our developers who will provide technical support. In order to execute a decision, 2 out of the 3 multisig key holders will have to sign.

The multisig will control the liquidation proceed parameters, which will be decided by the DAO. Despite an appearance of a conflict of interest, the security model is not being weakened. While a malicious backstop could nullify all proceed sharing, this is no different from a malicious backstop that simply stops liquidating (leaving the proceed sharing at 0).

Conflict of interest may arise if a malicious backstop refuses to on-board additional members, and in that event, the DAO would have the ability to revoke the multisig power by an on-chain voting.

It should be noted that in any period, the multisig will not have any control over the user deposited funds.

User incentives post April 26th

Further to the recent community discussion, users will accumulate both non- transferable score and transferable governance tokens. The non-transferable score will be used for the liquidation proceeds distribution (the Jar), and the token will only be used for voting.

If the initial liquidity mining program is approved, then in every block (for the first 3 months) roughly 0.47564688 BPRO and score will be minted in every block, and will be divided to users according to the proportional USD value of their collateral and debt.

For this purpose, in every block 4/5 of the block allocation will be given out according to the proportional debt value, and 1/5 will be given according to the proportional collateral value.

For example, if the protocol total debt is $50M, and a user has $1M debt, then in every block he will get (1/50 * 4/5 * 0.47564688) BPRO and score. Further, if we assume the total deposits are $100M and the user has $3M deposits, then he will additionally get (3/100 * 1/5 * 0.47564688) BPRO and score in every block.

EDIT: Discussion about Jar distribution post Genesis vote
Each protocol (namely, Maker and Compound) will have its own Jar, and a Jar will be distributed according to the non-transferable score whenever it holds more than $50k in value. Whenever a Jar is distributed, the user score is reset to zero.

The score and BPRO allocation, will be executed by the multisig on a weekly basis, and will use BadgerDAO implementation for that.

The allocation for the period between the beginning of the on-chain vote (April 26th) and the end of the vote will be retroactively distributed after the vote ends.

5 Likes

Perhaps SafeSnap can be of help here?

1 Like

We reviewed that solution, and talked a bit with the team.
SafeSnap security model relies on the ability to revert to full on-chain vote in the case of a dispute.
Letting the DAO replace the multisig in the case of malicious behavior gives similar model.
It is true that it might take some time to vote out the multisig. But unlike most DeFi projects, the multisig does not control user deposits (and neither does the DAO), so its attack surface is fairly limited.

Initially the multisig will have the keys to the DAO reservoir, but the DAO can ask the multisig the amount of BRO tokens there becomes a security threat.
This will not even require an on-chain vote. The DAO can vote to send the tokens to a contract that the DAO fully controls one in a while. And if the multisig refuses, then the DAO could vote it out, before the amount is too big.

We will continue to explore snapsafe-like solutions. Recently there are new projects who performs the full execution for you based on new crypto primitives. We talked with one such project, but their product is not live yet.

Whoops, missed the first thread completely. I have a couple of question and suggestions, just putting my thoughts out there as I’ve been keen on bprotocol since the very beginning as a potential solution to the imperfect on-chain liquidations that we see (gas wars/huge MEV front running opportunities that disincentive robust liquidations bots → could end up in a situation where the whole system is undercollateralized in an extreme case).

  1. Can we walk through the specific logic for the total amount of tokens? why 1,325m + 825k for perpetual minting over 4 years? Why 250k for a 3 month period and 150 for the genesis backstop?

That brings “max supply” to 10m. Any particular reason for this number, or the balance between the DAO reservoir and initial distribution and one time minting?

  1. I do feel like there should be a slight nod to maker users compared to comp users - they took more risk and jumped in when the project was completely new. Something like 52/48 or 53/47 seems fair to me - not a huge difference, but a nod to the earliest adopters.

  2. I feel very strongly that a 25%-75% split between collateral and debt for score accumulation is way too favorable to collateral. With such a high percentage, you will have ALOT of people depositing collateral with 0 loans, just for the sake of ‘farming’, and the vast majority of them will have no intention at all of ever taking out debt. I think something like 5%/95% or even 3%/97% is much more appropriate for a collateral/debt split. We want debt - getting people’s collateral in is just a form getting users to dip their toes in, and in this case I’m reasonably certain 5% is more than enough to incentivize plenty of collateral. At 25%, bprotocol will probably consist of 90% ‘dead’ collateral that’s there just to accumulate tokens with 0 debt. If you’re not taking any risk by taking out debt (or moving your existing debt over), your rewards should be minimal. Enough to get people to check the project out. Then you tempt them into taking out a loan with the high rewards for the user activity that we actually want to occur.

  3. BRO can have negative connotations and also is a bit weird. Why not BPRO? its not a big deal of course, but i think BPRO is much more naturally derivative of Bprotocol than BRO. If u have heard of bprotocol, if u see bpro u will know what the token is immediately, whereas if you saw BRO, it might be a ponzi-of-the-day on BSC or something. Definitely not the highest priority thing though.

  4. 2/3 multisig to begin with is fine, but i think there should be an effort to quickly on-board 2 more multisig members, with at least 1 from the ‘community’ (and maybe 1 well known figure who would suffer ‘reputational’ damage if they acted maliciously). 3/5 to me seems the right balance between efficiency and checks and balances.

Just my 2 cents

5 Likes

Is non-transferable score perpetual? Does it mean if you part with the governance token you can get the proportional profit from B.Protocol?

When I heard BRO for the first time, I thought it was good, but BRPO is more natural.

For me, 50:50 maker-comp proportion sounds fair enough because the amount of dollars in maker is less than comp. 50:50 maybe favorable already for maker users as early-adopters.

3 Likes

Thanks for the thorough reply.

Easiest answer to answer is regarding the 3 month period. The rationale here is let the DAO decide on the desired amount based on some empirical data. And after 3 months period it will be easier to estimate the affect.
250k means that if it will continue at that speed, then after 1 year, new users will get same power as existing ones, as additional 1M will be distributed to the new ones (which of course include existing ones). But of course it is hard to predict what will be the speed after 3 months (the DAO will have to decide).

Given the relatively low amount of liquidations so far, the backstop needs some incentive to continue, and having the ability to take some part of the governance could achieve that.
Recall that other than MEV, the protocol was originally deployed also to make the liquidators community stronger, with a more transparent way to incentives them.
The hope is that after 1 year a self sustained community of liquidators would be formed, maybe even user based.
As for why 150k, given there are 3 liquidators, each will get (after 1 year), 50k.
There is no right or wrong here, talking with them in the past, such request came from their side, but it is finally the DAO decision who will have to approve it.

Finally, the 825k per year that will be distributed among devs seems to match what other projects are doing, so we put it as the basis for discussion.

Here the rational was that Genesis governance users will have majority in the DAO roughly until the end of the first year.

I tend to agree, but there are two things to notice, (1) 50/50 already favors towards Maker. Compound tvl is over 30% higher at this point. (2) In the on chain vote, both Compound cScore holders and Maker mScore holders vote separately, meaning each could fail the vote. Hence, 50/50 would probably be easier to get consensus around.

Again, tend to agree, but note that there is also a user acquisition aspect here. Other members of the community suggested 20% and 28%, so 25% is kind of a balance of the two.
It should be noted that this ratio could be changed at any time with a vote. So if there are many dead accounts the community can vote to change the ratio. And the initial 1:3 ratio guarantees the dead accounts won’t get majority.

BPRO does sound better. @sunny any comments on that?

We will be happy to on-board members of the community.

3 Likes

In the discussed scheme, the token is only for voting.
And proceed sharing is according to score.
The score will reset every time the jar is distributed. This is at least what was discussed, but can be changed at any time.

I would like to emphasize what @yaron mentioned regarding user acquisition - As the DeFi market moves very fast, I think it can have great impact and value to have B.Protocol higher on the different “Leaderboards” the ecosystem got used to in order to attract new users - and the leaderboards are ranked by TVL.
So in this regard, having users with zero debt in this initial phase is not valueless, and might be one of the main growth engines for the protocol. After the first liquidity mining program ends, this ratio of course can be changed by the DAO - but then it will be much easier to make already active users start borrowing to keep on getting rewards than to bring new users to join and borrow.
As a long-term approach IMO, it should be considered to adjust this incentive scheme after B.Protocol is more mature/known/used.

My 2 BPRO…

1 Like

Frankly, and with no disrespect to @sunny, with the amount of scam-coins in this market, BRO does not give a good first impression, in my opinion. I agree with @amplice and @k_t that BPRO is much better.

I’m in favor of the 80:20 ratio proposed by @Dermot in the previous discussion. We will have a positive feedback loop (“virtuous cycle”) by strongly incentivizing borrowing on the platform. Relatively high liquidity mining rewards for borrowing will distinguish b.protocol and attract borrowers, more borrowers will result in more liquidations, more liquidations will increase the value of m/cScore, a more valuable m/cScore will increase the value of the liquidity mining rewards which will attract more borrowers, and so on. The governance token will be fundamentally more valuable as the jar distributions increase. We may even want to discuss a ratio that favors borrowing further.

I agree with @Yaron, and think it would be hard to get anything other than a 50/50 token allocation between Compound and Maker users through in a vote. Since Maker has had less value locked than Compound, giving 50% to Maker users, despite Compound users’ outsized allocation to protocol locked value, is a nice way to reward Maker users, many of whom were among the earliest adopters.

@Yaron, I don’t think it makes sense to give 150k to the liquidators. Given holders of the governance token will vote to decide who the liquidators are, giving the current liquidators voting power doesn’t seem right. Additionally, I don’t think ROI of 150k to the current liquidators would be better than spending the same amount on user acquisition via liquidity mining rewards, etc.

Lastly, I respect the complexities involved, but we should consider an integration with a low-fee protocol. Ethereum Layer 2/scaling solutions are seeing a lot of value capture given high gas fees on layer 1. TVL is growing much faster on Layer 2 than Layer 1, and there aren’t too many low-fee applications available to capture this value. This would be opportunistic given the current environment, but eventually the low-fee DApp space will become saturated. A lower barrier to entry would also diversify the user-base, increase the number of stakeholders, and therefore, reduce systemic protocol risk.

4 Likes

Ok, I will edit the post and change the name to BPRO.

As it seems most community members are in favor of worsening the ratio, it will be changed to 80:20.

In terms of governance power, they will never be over 6%.
And I think not giving them any power will just make everything more binary. Meaning their only way to speak out what is on their mind will be to leave. and given current liquidation volume is pretty low, there will probably not be much of a demand to take their place in the near future.

are you aware of any L2 lending platforms on Ethereum?
Some community members told me they are looking at BSC.

I am only aware of Aave on Polygon’s PoS side-chain, and have been following the metrics closely. L2 TVL has been explosive since Aave launched (from ~$10mm to ~$650mm in just the last week) which seems to be strong evidence for the marketability of near-zero-fee lending, borrowing, and liquidity mining.

There are now multiple Ethereum L2 solutions implemented with DEXs, but I’m only aware of Aave for lending on Ethereum. If we integrate with BSC, we should develop a second UI, as Ethereum maximalists will be displeased to see BSC-related protocols in the menu.

1 Like

@Yaron I think the distribution to the genesis backstop is entirely reasonable given early investment/support, but allocation amount is quite significant, and the net result is zero-sum dilution to the current c/mScore holders without any real alignment of incentives, as it stands. If we want to align incentives between b.protocol users and the genesis backstop, as b.protocol distributes 150k of BPRO to the genesis backstop, 37.5k worth of PBRO in KNC sponsored by Kyber Network and Onebit Quant could be distributed to the BPRO holders proportional to each individual’s BPRO share. This would also add additional value to liquidity mining BPRO and incentivize engaging with the protocol as early as possible helping to bootstrap adoption, while aligning b.protocol and Kyber Network simultaneously.

KNC is also listed on Aave V2 and would complement the eventual Aave integration, in a liquidity mining context.

Thoughts?

Why do you think this is significant? It is 0.5% per liquidator. And how else do will we get liquidators otherwise?

Asking them to buy their way into the protocol is a great idea. But I think we can get much higher price once we actually have many users and liquidations.
@yaron did other liquidators approached the team to join B.Protocol?

It is >2% per liquidator of all tokens created at the end of 1 year, 4% between the two. You can’t use fully diluted metrics, because the supply can be expanded indefinitely, given governance leeway.

Would the liquidators disagree with kicking-back 25% of the value of their allocation in KNC to better align the incentives of all entities, moving the total net amount allocated to the liquidators at the 1 year mark to ~3.2%.

The 25% value returned in the liquidators’ protocols’ tokens would help bootstrap liquidity as it would be an added incentive for liquidity miners. Additionally, in this case, the liquidators are not “buying in” to anything, as they are not entitled to 150k until/unless it is voted on first; I’m effectively proposing liquidators net 75% of the original proposed allocation, and 25% is converted to KNC and simultaneously distributed to the BPRO holders to align incentives.

no

Not sure why KNC is related here. 1 of the liquidators is kyber reserve, another one is market making on kyber, but also elsewhere, and another one was market making there, but currently does not.

Trying to collaborate with kyber dao somehow could be an interesting angle. But i do not see how it can be done before B.Protocol has a live DAO that can make specific decisions and proposals.

In terms on alignment with kyber, it is not a secret that it seeded Smart Future Labs, the company that developed B.Protocol in return for equity.
But kyber cannot take decisions for the kyber DAO (or KNC) and Smart Future Labs cannot take decisions for the future B.Protocol DAO (or regarding BPRO).

I agree that total supply is not the metric, as there is there is no supply cap. But not sure if it makes the impact of the 1st year allocation bigger or smaller.

One last comment is that I think that over at KeeperDAO the keepers who operate the bots get something like 33% of newly minted governance.

2 Likes

It is a cool idea to get some hard currency for the DAO.
But how can we decide it now? The BPRO is litterly priceless right now :slight_smile:

Also from what @yaron says only one liquidator is part of kyber.
The liquidators made very little profit in the last 6 months from us. I don’t think we have leverage on them.

2 Likes

This argument implies we really had no choice but to tokenize governance because the liquidators would leave without the added incentive. If such is the case, why was this not a focal point of the original tokenization discussion? It was not mentioned.

However, I did not account for the third liquidator (my apologies) and the 150k allocation is more reasonable among three entities.

@dr.sushi is not a member of the founding team. (This is to say he should not be blamed for raising or not raising issues, not that his opinion/voice does not matter.)

I do not think it is true.
The original protocol was design for a non transferable user based governance that will form a triangle of 3 entities, namely, users, devs and liquidators.
I do not agree with @dr.sushi that the liquidators do not have an incentive to participate.
I do think that at this point there is no leverage that can be used to ask the to pay for the participation (beside what they are already paying, namely, agreeing to share their proceeds).

I think that one of the pitfalls of existing DeFi protocols is that the voice of the liquidators is not heard. And I think the community letting them into the governance of B.Protocol would solve that.

Sorry for not being clear on that, i will edit the post to clarify that.

BSC is centralized, although temporary is certainly a better solution than paying an Eth gas fee.
I think Polygon is the best solution

2 Likes

Thank you for the reply and your super-clear editing in the proposal explanation.

Could you let us know more about the upside of not having the hard-cap in BPRO? Considering other protocols like KeeperDAO?

1 Like