Hey @Zer0dot, glad you like the idea and really appreciate the thoughtful questions. Will take them in turn.
1. Safety module
- As the protocol expands, the cashflows accruing to the ecosystem reserve increase. However, these necessarily increase slower than the cashflows to the specific aDAOs, particularly the higher growth ones since the ER receives less cashflows (20% vs 80%) and has a much higher numerator. Also, this doesn’t necessarily mean higher yield for $stkAAVE holders as we don’t believe all cashflows should automatically be paid out to holders but rather be used to build up a reserve / invest into new initiatives as per this proposal. The yield to $stkAAVE holders should always be the lowest available in the ecosystem because $stkAAVE holders aren’t backstopping any risk. Capital which wants higher returns will flow to the aDAOs based on their risk policies and mandates.
2. Liquidity Thinning
I’m not sure if you’re referring to liquidity thinning for the specific money market deposit tokens or for the $aDAO tokens so I’ll address both.
In the case of money market tokens, the proposal doesn’t actually fragment liquidity in money market tokens since each aDAO is granted an exclusive license over assets deposited into its protocol as well as the type of lending products it provides. The aDAO is kept in check by the fact that Aave can revoke its license if it feels like it’s not doing a good job. We are still on the fence about whether eventually different aDAOs can compete on the same assets, but right now we’re leaning towards thinking that the tradeoff in terms of fragmenting liquidity is not worth the added benefit of encouraging more competition to keep aDAOs in check. Especially since every aDAO will face both external competition as well as the possibility of Aave revoking the aDAO’s license, both of which should act to encourage performance.
In the case of $aDAO tokens, the idea is that aDAOs have their own differentiated mandates and resultant risk-return profiles. An aDAO isn’t limited to the tokens/products it initially launches with but will rather seek to continuously add tokens/products that fit with its mandate in order to keep growing.
I don’t see “liquidity thinning” being an issue because a new aDAO should only be launched if there is a ‘gap in the market’ in that no other aDAOs are adequately fulfilling its mandate. While some liquidity may flow from old aDAOs to new aDAOs, this should be a net positive for the ecosystem since it will increase capital efficiency by opening up a greater variety of risk-return opportunities for $AAVE holders and for investors who want exposure to the decentralised credit space.
On the other hand, I think there may be some fighting around which aDAO has “jurisdiction” over a certain asset, especially over time as assets which are initially risky and belong in a given aDAO may become less risky over time such that they belong in another aDAO’s mandate. Relatedly More research needs to be done on this but any decision of this sort should ultimately be decided on by $AAVE holders for the good of the broader Aave ecosystem.
3. Fee Distribution
At a high-level, Aave DAO would no longer be operating money markets but rather governing the aDAOs who themselves operate money-markets. As such, it makes sense that the aDAOs themselves build up treasuries since they will be closest to understanding the needs of their own money-markets. The alternative is extreme inefficiency and bureaucracy with each aDAO spending decision needing to be a proposal approved by $AAVE holders who are not best positioned to understand that aDAO’s needs.
It’s also worth noting that the Aave DAO should still be able to build up a significant capital base via fee flows from the various aDAOs as well as its ownership of $aDAO tokens. In any case, we really need to model this and tinker with it to see what it actually looks like. These parameters aren’t set in stone and are subject to being changed based on the modelling work to be conducted in Phase 2.
4. Bonding Curve
The bonding curve is necessary to incentivise boostrapping of higher-risk, early stage aDAOs by appropriately rewarding early believers. Otherwise, there’s no advantage to taking the risk early since later entrants can simply wait and pay the same price to stake in later once risk is lower / fees are higher. While a bonding curve isn’t the only way to do this, there needs to be some equivalent of a pricing mechanism to incentivise early adopters. For instance, while Nexus Mutual doesn’t use a bonding curve for staking on smart contracts, there is a functionally equivalent price curve which rewards early stakers:
$aDAO tokenholders are definitely taking a risk, but this unavoidable in other to create incentives for price discovery as mentioned above.
5. aDAO Governance
As context, there is a minimum quorum for proposed aDAO votes (we suggested 5%) and, provided the vote passes, $AAVE holders that vote for an aDAO will automatically be converted to $aDAO tokens. There are thus several benefits to launching an aDAO vs launching a competitor:
- Access to Aave’s existing tokenholder base and associated liquidity. This initial capital base allows an aDAO to launch with insured money-markets from the get-go, providing a better product for users.
- Access to a knowledgable, long-term oriented community to help govern and grow the aDAO from the beginning
- Access to Aave’s existing integrations, brand and traffic as all aDAO money-markets are listed on Aave.com’s front-end
As a metaphor, an $aDAO can be seen as an Aave franchisee responsible for certain markets/mandates. They benefit from Aave’s brand, infrastructure and know-how, giving away some of the upside in return. The tradeoffs of starting an aDAO vs a competitor are similar in character to the tradeoffs of starting your own burger joint vs buying a McDonalds franchise. In the digital world however, network effects mean the benefits Aave can offer are much greater.
**6. Reputation & Voting **
I would argue some of the same results are achieved, with the difference being a voter delegation system still assumes all tokenholders are involved enough in the governance process to know who it makes sense to delegate their vote to (kind of like the traditional political system). A reputation system, because it starts off centralized, puts most of the onus of initial distribution on the DAO’s team and is more likely to trickle through and end up in the hands of valuable community members.
I’m not sure if I understand your point regarding locking tokens with the FDS system. The idea of Future Days Staked is that it’s a voluntary commitment a tokenholder can make to lock his tokens for X amount of time in order to augment his voting power proportionately. Ultimately, it serves to provide greater influence to less wealthy but long-term convicted $AAVE holders than to more wealthy speculators, rewarding commitment as well as capital.
Reputation is subject to being gamed just as all systems are. The key is to create strong norms around distribution and what constitutes a valuable non-monetary contribution. Regardless, reputation is not the focus of the proposal and will be a later workstream once we’ve agreed on the high-level architecture of the system.
Conclusion
To address your last point, we don’t see this proposal as a way to circumvent security in favor or rapid innovation, but more as a way to enable secure innovation; reducing the potential downside of failed innovation and thus enabling more of it to occur. In all startups, it’s important to shorten the evolutionary cycle as much as possible, making small experiments, iterating and failing quickly in order to progress. This has arguably been Aave’s greatest strength so far and our fear with the concentrated power/capital model is that over time Aave’s size and the potential for contagion will impact its ability to do this.
Overall, really appreciate your contribution @zer0dot . You asked some excellent questions and this is exactly the kind of feedback we need to force us to think deeply and strengthen all aspects of the design. Hope the answers were alright and please feel free to follow up with any more comments, questions or feedback. We’ll make sure we get back to you asap!