The path to a unified governance

On April 26th, B.Protocol users will vote on a new governance and jar distribution model. Given the clear signal from the community, with 99% of the users wishing to base the governance on a transferable token, it is the time to discuss how a single DAO would control two different integrations, namely the Maker and the Compound integration.

The current governance consists of two DAOs, one is controlled by mScore (Maker) and one by cScore (Compound). In a tokenized unified governance, the DAO role is three-fold:

  1. Select the backstop members (liquidators).
  2. Decide on the proceed sharing scheme between users and backstop (jar distribution).
  3. Making the governance more decentralized, by further distributing governance power to users (a.k.a liquidity mining).

Building the unified governance will require to decide how much power each mScore and cScore holder will get in the new unified DAO. Or in other words, the initial token distribution between Maker and Compound users.

And how the future score will be distributed between Maker and Compound users.

Initial governance distribution

At the time of writing, the amount of Compound deposits (in B.Protocol) exceeds those of Maker by over 40%, and we are witnessing a bigger growth there. However, Maker users were the first to join B.Protocol, and had a major role in creating the initial traction around it.
In the absence of a mathematical formula to weigh such different considerations, we hope the community could form a consensus around splitting the initial power equally, namely 50% to Maker’s users, and 50% to Compound’s.

Future governance allocation and Jar distribution

Moving forward, in a unified governance, a liquidity mining program will have to be able to weigh operations across different tokens (e.g., WBTC, ETH, DAI) and different platforms (Maker and Compound).

Members of the community (e.g., @Commiekiller) already proposed a scheme in which everything will be normalized to a USD value. We believe that this is an elegant way to balance operation in different tokens.

In order to balance between Maker and Compound platforms, we need to take into consideration that normal Compound users sometimes only deposit their funds, and do not borrow, while a normal Maker user does not. While our backstop doesn’t generate added value for a user that only deposits funds, at these early stages it would make sense to aim for all Compound users, with the hope that once they are using B.Protocol interface, they will continue using it also when they wish to borrow.

To mitigate these differences, new Score and governance rights will be distributed according to the USD value, where the ratio between Score allocated to collateral and debt will be 1:2.5. Meaning a debt will get x2.5 over deposits. For fairness reasons, we suggest that the same scheme will be applied also for Maker, even though deposits serve a different purpose there.

Finally, because of different user profiles, some protocols will have more liquidations than the others. As the community already suggested to distribute the Jar according to a non transferable score (and thus, decouple it from the voting rights), it would be appropriate to maintain different jars for each protocol, also in the future, and distribute each Jar according to the Score that was accumulated for that Jar by the users of that platform.

To put in simple words, the separation to mScore and cScore will remain, and the mJar (respectively, cJar) will be distributed according to the mScore (cScore), while the governance power will be unified for both integrations.

Summary

  1. Post Genesis vote, 50% of governance will go to Maker users, and 50% to Compound users, and one DAO will be formed.
  2. Future allocation will be according to a normalized USD value.
  3. Deposits, both in Maker and Compound, will be rewarded, according to 1:2.5 ratio compared to debt .

We invite the community’s feedback on this scheme in the coming days, before the on-chain vote will begin.

6 Likes

Does it make sense to split the governance score 50/50? It’s much easier to accumulate score in Compound since it gives credit for both supplying and borrowing. mScore is harder to accumulate because you only get credit for your debt. Splitting 50/50 seems to favor new Compound users over early Maker participants.

My cScore will exceed my mScore by the genesis vote, despite supplying and borrowing half the amount for 1/5 of the time.

This is a very valid point and one that should be considered. There are 2 points of score distribution on compound, lend and borrow; whereas there is only 1 point of score distribution on maker: borrow.

To further compound this inequity, folks can actively mine comp by recursively lending, borrowing, lending same asset again, borrowing again, etc. I’ve seen some folks recursively loop this process around 10 cycles or so. So in reality, they are accruing an extremely large cScore with relatively lower assets, and making significant income from comp mining in the process.

So I’m not sure it’s equitable to compare mScore to cScore. We need to have a serious discussion about all these points and come to a conclusion that doesn’t make the makerDAO users feel like second class citizens in bprotocol. A 50/50 split could be contentious with makerDAO users feeling taken advantage of.

Also, makerDAO users were the first to support bprotocol, being a brand new protocol, they exposed their capital to much more risk than compound users, who only started using the protocol after it was battle tested by maker users. I think it would be a mistake to not factor this into the governance split equation.

Considering all of the above, I provide a suggestion, which I’m willing to modify in light of any other evidence, that rather than a 50/50 maker/compound split, it be 66.6/33.3 maker/compound. We’re all the bprotocol team and this shouldn’t be a contentious issue, and we should strive to do what’s best for all participants, of course there will a minority whose only priority is to maximize their own governance distribution, but I think most people here are willing to support a solution that is the most fair. I’d love any feedback on this idea. FYI I’m in the comp vault, so this wouldn’t necessarily be in my best interest, but I think it’s the most fair solution and in the best interest of the protocol and users.

Edit: I think the deposit/borrow weights of 1/2.5 is extremely fair and support it fully. If this were to be applied retroactively such that maker users score was adjusted based on historical deposits prior to governance distribution, perhaps 66.6/33.3 could be adjusted downwards to something like 57.5/42.5; based on the aforementioned inequities.

I think you are misunderstanding the 50/50 split. It gives mScore holders an outsized allocation of governance power, with respect to value locked in b.protocol.

3 Likes

My understanding of the split 50/50 as a Maker user,i will receive close to 50% more governance power as opposed to if we went for a straight 1:1 M score and C-score.Maker user comes out fair imo.

I think the fact that it is hard to compare the scores strengthen the 50/50 argument.
50/50 means that if a user has 10% of Maker total score, and 20% of compound total score, then he will get (10% + 20%)/2. So the fact that it is easier to accumulate score on compound is mitigated by looking only at the relative % of total score in each protocol.

  • please let me know if this answer your concern, I am not sure I understood it.
2 Likes

In general the 50/50 does favor Maker users (over just taking mScore = cScore).
It is true that it is (currently) easier to accumulate score on Compound.
But a 50/50 (or any X/Y) scheme fully mitigates it. As 50% goes to Maker users, regardless to how many scores/deposits there are on Compound.

In that regard, retroactively adjusting the score of Maker will not make any difference, as Maker users will still get 50%.

I also feel that Maker users should be rewarded for being first and longest time.
50/50 already does that, but can understand why you might think 57/43 scheme might be more appropriate.
That said, we should recall that eventually Maker and Compound users should each separately approve it. Hence, 50/50 might be easier to get consensus around.

3 Likes

Thanks @yaron, for this thoughtful post.

It seems like there are two questions to resolve:

  1. How to value contributions up until the initial governance distribution?
  2. How to value contributions following the initial governance distribution?

And to resolve these two questions, I have another question (!) - what is the north star metric for B Protocol? Is it the total $ of borrow managed by the protocol? Or something else?

Working off the assumption that it is total $ of borrow managed by the protocol, I suggest allocating 80% of weight towards that and the remaining 20% towards other contributions that support the north star, most obviously starting with deposits and leaving it open for governance to decide future contributions to add to this bucket in the future.

Dealing with my initial questions:

How to value contributions up until the initial governance distribution
Subject to what I said above, it seems fair to allocate initial governance holdings 50/50 across contributions made via Maker and Compound. Even though Compound accounts for a larger proportion of borrowed $, Maker was the first protocol supported, and the protocol should reward early backers for such support. You could use a simple weighting to illustrate this, but I imagine it will end up looking like something close to 50/50 in any case.

How to value contributions after the initial governance distribution

Following up on my initial point around the north star metric for B Protocol, should it not be the case that all $$ of borrow managed by B Protocol is treated the same across all of the future protocols it supports? This should alleviate scaling issues around how to treat borrowed on new protocols added like Aave. This gets around the tribalism of where the is coming from and aligns all governance token holders on growing the total amount of $ of borrow managed by B Protocol.

Keen to hear everyone’s thoughts.

2 Likes

I think this is fair proposal. I dont understand why 50:50 is not fair to MakerDAO users? The only fairer solution than 50:50 is to treat mScore and cScore equally and tokenize their sum - (mScore + cScore = tScore).
My opinion is that bProtocol should not adapt to cryptoeconomic mechanisms and incentives of underlying protocols.
I agree with @Dermot, we need define NSM(s) for the purpose of making objective decisions.

Just to note that I voted against tokenized governance solely because the mechanisms of each underlying protocol are different. If the token distribution will adapt to the way MakerDAO works, then this will need to be done with each new protocol that is enabled (i.e. Aave, bZx …)

1 Like

Great replies guys. I had a fundamental misunderstanding about the nature of the split, and I retract my suggestion 66.6/33.3. I think 50/50 is acceptable, but maybe we should bounce around the idea of a borderline trivial split like 51/49 or 52/48; just to reward the protocols earliest users who took the most risk.

1 Like

Moving to tokenized governance give rise to two questions, what is the north start metric for governance and for users.

I think user-wise the goal is simple (though hard to achieve…): Maker (respectively Compound) users should see it is an easy decision/no brainier to user Maker (Compound) via B.Protocol interface.
This means that new Maker (Compound) users will open their new accounts with B.Protocol, and existing ones will import them.

Ideally trying to target mostly the users that might become a systematic risk for lending platforms (e.g., big accounts with high debt) would serve the purpose better, but it is harder to get right without making it gamable.

So rewarding mostly debt fits well to the ideal target, while giving some reward to collateral fits well with the user acquisition goal.

This means 1:4 ratio (as opposed to 1:2.5). This is definitely doable, and would be interesting in hearing more thoughts around it.
It goes without saying that the DAO will have control over it and will be able to change it over time.

As this could be changed by the community over time, it is probably the easiest to start with just normalizing it to USD value and treat every USD equally.

Or alternatively, as suggested, adapt the way Compound works to MakerDAO, meaning give some reward to deposits, even though it is a bit artificial.

1 Like

Some symbolic gesture might be in order, but if eventually this prevents forming a consensus it might defeat the purpose.
And it should be noted that 50:50 is already a gesture towards Maker users.

1 Like

I believe part of this goal involves finely tuning the financial incentives of using B.Protocol, but the other piece is the user experience of interacting with it. In order to truly make it a no-brainer, the user needs to feel as if they are not making any concessions in usability or potential revenue. To me, that means:

  1. The UI for interacting with their funds on the platforms supported by B.Protocol should be as good (or better!) than the native interfaces.
  2. The DApps, tools and financial legos that are interfacing with the protocols we integrate with should exist for B.Protocol users, as well. (DeFiSaver, Zerion, Zapper, DeBank, etc.)

Ideas on achieving these and other goals would probably be suited for another thread, but ensuring we can focus on some of these areas would likely need to be part of the governance discussions at some point.

2 Likes

Thanks! Sorry, I misunderstood the initial proposal and I’m in agreement that everyone is fairly considered.

I read B. protocol’s twitter.
We should defeat Keeper DAO who is spelling burrow instead of borrow :slight_smile:
Let’s have more and more discussion for the future success.

As it seems the community agrees on most things in this post, the discussion would continue along with the full token model here.

2 Likes