Thanks for your early comments guys.
@0xd15co, I appreciate you bringing overcollateralization up. It’s a key component of eUSD’s architecture and should always be top of mind, especially in proposals like this where we are discussing fundamental changes to collateral approval and basket construction frameworks.
I broadly agree with the problem statement you’ve outlined. As the collateral universe expands and diversifies into the long tail, it is reasonable to ask whether a static view of “adequate OC” and stRSR revenue share remains sufficient or if these should vary based on the risk profile of the collateral basket.
Where I struggle is with the practical implementation. While a stress-adjusted OC framework sounds sensible in principle, defining potential stress losses, dependency failures and an appropriate stress-adjusted value for RSR quickly becomes highly complex. I worry that introducing such a framework could shift governance discussions towards the scrutiny of risk models rather than basket construction itself.
While much simpler, we’re already seeing this play out for ETHplus as differing interpretations of the ETHplus methodology are used both in support of and against the latest rebalance proposal. While this is healthy and expected, adding a mandated framework for evaluating basket risk may push us towards a level of governance complexity I’d rather avoid.
I also think that eUSD’s stRSR model already balances yield, risk and over-collateralization quite nicely. As, any increase to basket yield, increases RSR staker revenues. In theory, this should encourage additional staking and increase the size of the staking layer over time. At the same time, stakers provide first-loss capital. If we assume higher-yielding baskets generally carry greater risk, this creates a natural balancing mechanism where stakers are incentivised to balance yield generation against collateral risk.
That said, I don’t think the relationship is straightforward enough to rely on. Markets are often inefficient, staking participation doesn’t always respond proportionally to yield and there may be a significant lag between increased revenue generation and increased OC. While basket risk, protocol revenue and OC are clearly related, the strength and timing of these relationships are difficult to quantify.
There is also a broader product design consideration around predictability. One of the objectives of this RFC is to provide integrators, fintech partners and regulators with greater clarity around how the basket is managed. eUSD revenue forms a core component of UGLYCASH’s business model and future fintech integrations are likely to place similar importance on predictable yield generation. A dynamic relationship between basket risk, OC requirements and stRSR revenue distribution may improve risk alignment, but it would also reduce product predictability and weaken the incentive for fintechs to build around eUSD.
For those reasons I kept the methodology relatively high level on OC requirements. My view is that this RFC should focus on establishing a framework for collateral approval and portfolio construction, while preserving governor discretion when evaluating whether OC remains appropriate for a given basket composition.
That said, I do think you’ve identified an area that deserves further discussion. Competitive risk-adjusted yield should not apply solely to fintech customers, but also to RSR stakers who are ultimately underwriting the basket. As governors evaluate future basket compositions, the relationship between basket risk, protocol revenue, staker compensation and OC should remain an important consideration.
@blue, 100% agree! A 50% allocation into a single protocol feels right today but this can definitely come down as the collateral universe expands. Another consideration is to limit allocation by yield type; pooled lending, curated lending, RWAs, DEX LPs etc makes a lot of sense to me as the collateral universe matures. Only thing I think we need to be careful of is tightening these constraints too early. For example during market stress we may need to reduce collateral baskets risk, removing enhanced and opportunistic allocations or exit one protocol completely e.g Aave during the rsETH hack, if constraints are tight it may be impossible to redistribute allocation while still being aligned with the methodology. Although, this could be remedied with a clause in the mandate that allocation restrictions can be lifted during significant market stress. Food for thought for any upcoming methodology amendments…