Compound Finance - An Appetizer
It’s no secret that Compound Finance is one of the premier money legos in the DeFi space. The protocol serves as a two sided platform for supply and borrow markets, these markets in turn function as distribution mechanisms for the native COMP (governance) token. It now occupies a more foundational layer in DeFi; in Q2 of 2021 nearly 35% of the supply side came from a single protocol (Instadapp’s flashloan pool, Instapool). These protocols leverage the stability of Compound, a stable platform is needed at the bottom layer as more complex components are stacked on top of it. In that same vein, it’s much simpler and less risky for an individual to move liquidity elsewhere if they disagree with a protocol change relative to an entire complex system built on top of it. That is all to say, governance matters. Let’s now explore the Compound governance system.
Compound Finance - A Governance Entree
The Compound protocol as a whole is governed by COMP token holders. Token holders vote on proposals, successful proposals are moved into a queue called the Timelock where they wait to be executed after a 2-day waiting period. Proposals in Compound are executable code, which means there’s some technical overhead to even create a proposal but the change itself is accessible to all and the code can be reviewed by any interested party. Voting is held on-chain, bringing with it a gas cost to participate. As a COMP holder you’re free to participate (or not) in each proposal, additionally holders can delegate their voting power to any address. This governance system is generally codified in Governor Bravo. This is essentially the engine that drives governance efforts. In fact, it’s such a powerful governance tool that other protocols, such as Uniswap, have implemented versions of its predecessor, Governor Alpha.
Source: Compound Governance
Proposals can range from updating collateralization ratios, adding new assets to the platform, and the latest proposal reduces the amount of COMP needed to create proposals (now at 100k, would be reduced to 65k), including autonomous proposals. The threshold of wielding 100k COMP in order to create a proposal could be seen by some as too burdensome and restrictive to users. One solution to this issue was the creation of autonomous proposals; these are proposals that anyone with 100 COMP can create. Once the autonomous proposal, which is a smart contract, reaches a threshold of 100k COMP delegated to it, then it can be wrapped as a real-deal proposal and go through the process described above.
The Compound grants program, which was also a governance proposal, was allocated 5000 COMP to further encourage and incentivize protocol development. The program will run until September 2021, but who’s to say it won’t get extended by another governance proposal. This is the power token holders possess, they can recognize community member contributions by directing protocol capital their way.
The governance system described above was not created or developed overnight. To reach the current state of governance practices, the project has followed a progressive decentralization approach. What this means is that over time (as the protocol has ossified itself within the DeFi space) governance functions have shifted from being handled by a trusted set of team members, to holders of the COMP token. This is not a trivial process; it requires building a core product, developing a community around it, and a distribution mechanism for disbursing the governance token. Each of these bring about their own challenges, and one misstep can cause a huge setback to the entire process.
Compound Finance - Lunch Line Leftovers
We know how it works now, let’s check out some governance and voting data. Tally is an excellent governance tool, it’s got some great data on Compound governance. No peeking here, so take a guess; out of ~171k COMP holders, how many total addresses have voting power delegated to them? 20% of holders? 10%? 5%?? No, try ~1%! That’s a bit concentrated (for better or for worse).
According to Compounds leaderboard the top 5 addresses hold nearly ~43% of the total voting power. The average voter turnout over the last 10 votes is ~45 total addresses.
Having a small set or committee of people govern on behalf of a much larger population isn’t a novel idea, obviously. So maybe it shouldn’t be too surprising to see this pattern emerge. At the same time, the increasing rate at which DAO’s keep popping up will inevitably stretch out the spectrum on which DAO’s sit. Consider a venture DAO that invests in early stage DeFi projects. The members of that DAO are there for a very singular purpose, so I’d expect to see a more robust voter set relative to Compound.
Nonetheless, delegation is an excellent mechanism to allow holders to pass through voting power to other community members while still keeping exposure to the underlying token. Keep in mind, nothing really forces a token holder to delegate their voting power to another address but clearly holders are doing this. People choose to use this in order to effectively delegate their decision making authority. That’s a key but subtle point; people actively pass on their role in reaching consensus to others who (I’m assuming and so are those delegators) are aligned with their interests.
There will naturally be debates as to how decentralized the governance process really is since power can work its way through a funnel down to just a few addresses. These debates are very much needed, and protocols should do their best to be transparent about this. But, reaching consensus on protocol changes is a key objective of governance. That’s very difficult to achieve in an extremely flat, no-delegation, environment. So, how is a balance struck between general holders of a token (passive holders) and the folks who are really in-the-know about what needs to get done (core designers/devs, etc.)?
Many protocols are facing this challenge as their scale increases, and my hope is to see exploration in the design space around this challenge. One of the leaders here is Yearn Finance and their governance 2.0 model. This proposal really deserves its own piece, so I will get to that at a later time. The method of “constrained delegation” proposed by Yearn is a natural extension of the direct delegation concept that protocols like Compound have.
As DAO’s continue to scale, especially in DeFi, we’ll start seeing more of this model. Things move so fast in this space, but we have to remember that it’s still very early here. There’s so much that needs to be built out: What tools do these organizations use to initially distribute tokens? From a financial perspective, how do they responsibly manage treasuries? When should a centralized protocol make a move to meaningfully decentralize? What tools can be used to reduce friction when paying out salaries/grants/etc.? How do we fairly value community member contributions? How are tensions relieved between all the different sets of stakeholders? How do DAO’s interact with each other?
All these questions are what really excite me about this space. We’re rethinking how we organize and coordinate with each other at a fundamental level, no doubt this will lead to fascinating outcomes.