Amend the ragequit mechanism to use a hardcoded Intrinsic Value rate, set 1 hour before the DAO vote begins. This will use the same calculation of Protocol equity / circulating TRIBE as reported by collateralization oracle.
This mitigates concerns around oracle manipulation during the window, and improves the game theory by removing volatility and perverse incentives.
Are we expecting that there will be users who will take ragequit as an immense arbitrage opportunity (30%+) and is this positive behavior for the protocol? I can see some people using ragequit to redeem their TRIBE at the intrinsic valuation and then dumping it, sinking the price and then re-entering the market.
If so, I would like to propose an amendment in the ragequit contract to prevent arbitrage exploit. I would suggest to add in a cooldown timelock (maybe 7-30 days) as an escrow, where all of the TRIBE is āpreswappedā back to FEI, but that FEI is not redeemable until after the cooldown period.
It is an immense arbitrage. Think of it as both a reward to pre-existing holders and a huge buyback below IV. All the bought back TRIBE goes into the treasury to help fund future development.
The main reason to cooldown the buyback is to prevent a bunch of FEI from being redeemed all at once, which is a legitimate reason.
Arenāt we allowing TRIBE to be redeemed for IV, not below it? Also, does this amendment still use the snapshot window that GFX uses for eligible TRIBE?
Yeah it will still use that GFX snapshot block.
The redemptions would be at IV that is true. The arb goes to TRIBE holders who held TRIBE at the time (they couldnāt have known about the arb in advance)
Here is a snapshot for this amendment: Snapshot
Is the additional timelock proposal accepted and going to be incorporated into the snapshot? I currently do not see it there.
Not this proposal I think it adds too much complexity but feel free to put up a similar snapshot. Iād either abstain or vote no
I think it would be better to just ensure that there is enough ETH standing by to redeem the FEI coming from ragequitters.
As soon as people expect the vote to pass, TRIBE holders who are eligible for ragequit will start buying TRIBE and RGT (by the way, this is why RGT holders will benefit from ragequit the same way as would non-eligible TRIBE holders). Ragequit FEI wonāt be distributed for at least a few days after the merger passes, so these buyers need up front capital anyway. And even if FEI distribution is delayed, the spike in FEI supply will happen at the time of distribution unless some ragequitters get their FEI first.
Maybe Iām missing something, but only the TRIBE held at snapshot block is valid for RQ. RGT Holders wouldnāt benefit, nor would people who buy new TRIBE.
How much arb happens depends on who wants to actually exit. If the eligible TRIBE holders all want to exit, then they will just ragequit with the TRIBE that they already have, and that would be it with no arbitrage happening.
But if the eligible TRIBE holders want to keep their TRIBE, and the ineligible TRIBE holders and RGT holders want to exit at IV, that leads to arbitrage. The eligible TRIBE holders will buy TRIBE and RGT from the ineligible investors and ragequit for them. There is one group of investors who have the right to ragequit but have no use for it, and another group who want to ragequit but donāt have the right, so there will be a massive transfer of TRIBE and RGT from the latter to the former.
Having a liquid FEI-TRIBE pool will facilitate this transfer, although people could also trade on CEX of course. This was one of the reasons why I thought IDO should be maintained until the merger.
@joey I appreciate your leading the charge to improve the rage quit mechanics, but I still have major concerns with the rage quit mechanics. The first two are quite serious - and seem fatal.
Using the pre-deal numbers incentivizes everyone to rage quit and amounts to a 51% attack by old TRIBE holders on new TRIBE holders
Limiting rage quit eligibility further drives this dynamic
Using pre-vote prices does not improve the game theory
Letās unpack these one at a time. Then letās determine the best path forward.
1. The rage quit mechanic is effectively enables a 51% attack by old TRIBE holders on new TRIBE holders who will be drained of their share of protocol equity
As currently proposed, the rage quit offers all pre-snapshot TRIBE holders the chance to cash for the pre-deal pro-rata share of the protocol equity. However, an additional ~350m TRIBE will be issued in the merger. This will immediately reduce the protocol equity per TRIBE by about 1/3. In addition, some holders of TRIBE will have purchased it after the eligibility date.
HENCE EVERY OLD TRIBE HOLDER SHOULD RAGE QUIT and buy back in.
Moreover, because the protocol will be paying rage quitters a 30+% premium to the current protocol equity per TRIBE, the protocol equity per remaining TRIBE will go down even more. This will create even more incentive for āold TRIBEā holders to rage quit, leaving even less protocol equity for ānew TRIBEā holders, including the incoming holders of RGT.
Put differently, we are creating a 51% attack on the protocol in which old TRIBE holders transfer wealth to themselves from the PCV at a cost to new and future TRIBE holders ā e.g., those who bought after the snapshot; those who currently only hold RGT.
Is this really what we want to do?
2. Limiting eligibility creates 51% attack-like incentives
The other problem with the current proposal is that it limits the rage quit option to a select group of TRIBE holders. This again creates a ā51% attackā scenario where any ā51%ā of TRIBE holders has the incentive to set a criteria (like holding date) to take PCV from the smaller group of holders.
This will happen as long as not all TRIBE holders are eligible to rage quit.
Is this what we want to do?
3. Fixing the details doesnāt solve the bigger problem.
As described above, using pre-vote prices or changing other details of the rage quit will not change the incentives for some TRIBE holders to 51% attack the protocol and transfer wealth to themselves from the protocol at a cost to other TRIBE holders. To do that, we need to re-think the rage quit mechanic.
The Best Path Forward is to execute the merger and table the rage quit.
In light of these concerns, I suggest we remove the rage quit option until we can think through the implications more thoroughly. I believe it is especially important to table the rage quit now because we already have our hands full trying to perform the first DAO-to-DAO merger with RARI. Why risk derailing that to work thru the difficult issues here?
In addition, I donāt see any conceptual linkage between the rage quit and the merger except that we will be transferring protocol equity to ourselves at a cost to post-deal TRIBE holders. This does not seem like a way to quickly close the merger or win support from the Rari community or new holders of TRIBE.
CC @arcology
CC @Eswak @fei.saver
How much TRIBE circulating?
How many TRIBEs are currently owned by the DAO Fei Protocol?
You need to do a visualization of the TRIBE circulation, similar to the Protocol-owned FEI : Fei Protocol
This is a very nuanced discussion and Iām glad to see you call attention to these points.
I am in somewhat of a unique position having discussed the merger with many large TRIBE holders who bought on the market and have their own intentions and motives for holding TRIBE.
To them, TRIBE is at least partially simply the sum of its equity, and that is why they hold. The merge makes this value proposition 30% lower as you say.
The intention for the ragequit then is to allow all tribe holders unhappy with the new direction to exit with their pre-merge share of protocol equity
This naturally comes at a cost to the PCV but it has a few benefits:
In summary, the ragequit is simply an on-chain entitlement to what was fundamentally already belonging to TRIBE holders in a proportional manner.
I have reason to believe the merger would not pass on the TRIBE side without this feature. Feel free to post a related amendment about removing it to see how TRIBE holders feel.
Thanks for the additional context Joey. Since my post of a new topic seems to have ruffled some feathers, I wanted to start by saying my intent in raising this issue was to ask the hard questions so that the protocol is set up for long-term success.
I also understand the investor standpoint. That discount to protocol equity is one of the key reasons I have bought and held TRIBE, too.
Nonetheless, the mechanic will create winners and losers among the community and some of the results seem odd.
For example, do we really want holders who are happy about the new direction of the protocol (and therefor do not Rage Quit) to miss out on their current share of the Protocol Equity?
And what about you and the Team? Do you all plan to Rage Quit and re-buy back in?
And how about the VCs? Or the protocol controlled TRIBE?
I donāt think there are obvious answers here! But the incentives seem perverse and extend well beyond the merger wth RGT.
How are you, the team and the protocol planning to handle your TRIBE?
Canāt do governance right without ruffling feathers 
I think the spirit of your post to ask the right questions is great. Here are some answers from my personal perspective and feel free to dig deeper.
If there is an arbitrage opportunity, TRIBE holders who do not wish to exit permanently can take advantage of the arb and potentially end up with more TRIBE than they started with.
Agreed! But it is more fun when things go smoothly and we can buidl!
To make sure I understand this correctly, is this the correct analysis of which TRIBE that will be eligible for the Rage Quit?
40% DAO treasury ā Eligible if spent by the DAO (max: 5%?). Not eligible if held by the Treasury
20% Initial DEX Offering ā All 20% is eligible
13% Fei Core Team ā None eligible because none yet vested
10% Genesis Group ā All 10% eligible
10% Staking Rewards ā Eligible if distributed already (max 4%?)
5% Fei Core Team investors ā Not eligible because none yet vested
2% Grants Is there any TRIBE eligible beyond the Genesis TRIBE, public sale ā Eligible if distributed (max 1%?)
Non-circulating TRIBE is existing treasury (~300m) and staking rewards (~60m). Of the circulating TRIBE, (~300m) is timelocked in some form. This leaves (~340m) available as possible to ragequit. Assuming 100% of these ragequit there will still be several hundred million in protocol equity (and a huge treasury). This would be quite dramatic, hard to say what the real number will be but I expect it to be much lower than this.
Your arbitrage scenario wont work imho.