We should discuss the future of PoolTogether governance.
I don’t think an immediate decision needs to be made, but I think it’s worth starting the conversation. I’ll open it up with some observations about the current state of governance, then highlight some directions we could go.
State of Governance
- Gov no longer controls the protocol; it is fully immutable and autonomous! We no longer have any need to extend governance control across multiple chains.
- The POOL token can no longer be minted. The POOL tokens in the treasury are a finite resource.
- The treasury has no revenue. The non-POOL assets in the treasury are finite.
- A lot of POOL has been bridged to L2s (~25% of total supply). Bridged POOL cannot be used to govern
- The few voters on proposals are the old guard who’ve been around for a long time. They may not stick around forever; as they dwindle so will our defense against governance attacks.
Directions We Could Go
Below are some different directions I’ve thought of. Please comment if you think of another!
Option 1: Dissolve governance
We disburse the remaining funds in the treasury, finalize the LP positions, and renounce ownership of the Timelock.
Advantages
- This would fully decentralize PoolTogether. We may even be the first to do so. All that remains would be the token and the protocol
- It would not be possible to “attack” PoolTogether. We would no longer have a honeypot to attract governance attackers.
Disadvantages
- We’d need to disburse the POOL in a final distribution. We could build a way for incentives to be dynamic, but we wouldn’t be able to have discretionary spending (i.e. incentivizing a vault proactively)
- We wouldn’t have stables to draw on to fund frontends, builders or other initiatives that require resources.
Counterpoint: our treasury is limited anyway so it’ll run out at some point. Perhaps doing a final budget for each frontend is inevitable.
Option 2: Upgrade governance to support L2 voting
We’ve been told by Scopelift, who upgraded our Governor contract, that it may be possible to allow L2 token holders to vote.
Advantages
- All of the prizePOOL and POOL holders on L2s would be able to vote cheaply.
- The risk of a governance attack is far less, since all POOL would be eligible to vote and defend.
Disadvantages
- It’s unclear how much the capital and operating expense would be of the upgrade. When we simply upgraded the governor contract it was $50k: upgrading the contract again and adding L2 support would likely be much more up-front along with on-going costs.
- The treasury is finite; so voting would eventually end anyway. Is it worth it?
Option 3: Spin up governance on a single L2
Tally offers the ability to easily spin up governance system in a no-code way.
Advantages
- Voting would be cheap!
- If the system uses staked POOL as the gov token, then the L2 treasury could fill with ETH
Disadvantages
- We’d need to migrate all of the assets held by L1.
- We’d need to finalize or withdraw our PoL LP position
- We’d need to do the work to dissolve L1 governance
- Some work would be required to make staked POOL the gov token
- POOL on other L2s wouldn’t be eligible to vote. This means there is less active voting power, and governance would still be more vulnerable to manipulation (as it is now).
- The treasury is finite, so voting would end some day anyway
Option 3: Spin up governance on each L2
We could spin up governance on each L2 that we’re deployed on.
Advantages
- (same as option 2 above)
- People could vote on any chain
Disadvantages
- People could only vote on the chain they are on
- (same disadvantages as option 2 above)
- Coordination would be much harder. If someone needs budget greater than a single chain; it would have to draw from multiple proposals across chains.
Option 4: Create an off-chain entity to custody the funds
We could spin up a separate entity, such as the Marshall Islands non-profit, to hold the funds and spend them in support of the protocol.
Advantages
- Reduces the coordination required of voters, but still allows for it
- Eliminates the attack surface by having people with fiduciary duty manage the funds.
Disadvantages
- Costly to set up
- It’s a complex decision to choose what type of entity to use
- Someone needs to KYC and lead the entity and become a potential legal target (even if the entity is a liability shield).
- We’d still need to close down our L1 governance and transfer the assets to the entity.
My Thoughts
Personally, I believe Option 1 is best. The treasury is finite, so we shouldn’t spend resources and time coordinating on something that will eventually go away anyway. We lose the ability to offer discretionary funding, but gain deep simplicity in the protocol. That being said, if people want to combine their POOL staking returns and spend it on PoolTogether infrastructure, then they can spin up their own governance if they want to.
The steps we’ll need to take to dissolve L1 governance will have to happen anyway for the other options, so it’s the option with the minimum amount of work.
It’s also an opportunity to do what no other project has done; to fully eliminate governance and fully decentralize.
Your Thoughts
Please comment with your thoughts! I want to hear:
- What direction you like
- Whether there are other directions
- Whether there are more advantages or disadvantages to discuss