Summary
This proposal seeks to update the dgnETH collateral basket by migrating its 75% allocation from the Convex CryptoSwap ETH+/ETH LP to the StableSwap variant. The migration aligns the dgnETH basket with ETHplus’s own liquidity optimisations. The new StableSwap pool offers tighter spreads for pegged assets, reduced slippage on large trades, and improves ETHplus scalability. The StableSwap pool has recently delivered higher yields than the CryptoSwap pool (5.91% vs. 2.27% 30d average), though dilution is expected as liquidity migrates, mitigated in part by a subsequent migration of incentives by ABC Labs. In addition, the new plugin introduces gas optimisations that cuts dgnETH transaction costs by ~33%, albeit with a modest increase in stale-data risk as it moves the plugin uses 24h refresh window compared to the current plugin which refreshes on every action.
Problem Statement
Recently the Curve ETH+/ETH LP was updated, migrating from a CryptoSwap to a StableSwap pool. Given the dgnETH collateral basket still holds 75% of its basket in the Convex variant of the old CryptoSwap Pool, which is being decommissioned, this proposal aims to migrate the allocation to the new StableSwap variant.
Rational
Recently ABC Labs have undertaken work to optimise the efficiency of ETHplus liquidity on Curve, these optimisations involved migrating the LP from a CryptoSwap Pool to a StableSwap Pool. Given ETHplus and ETH are tightly correlated the new pool allows for larger trades with very little slippage when the pool is in balance. The change aims to support ETHplus scalability towards 9 figures.
With the creation of the new LP complete ABC Labs now intends to wind-down the incentives streamed to the old CryptoSwap Pool and migrate them over to the new pool once the dgnETH updates its collateral basket.
Basket Composition and Yield Comparison
Current Basket
| Collateral | 30d Yield (DeFiLlama) | Allocation Weight | Allocation Value ($) | Protocol Allocation / Collateral TVL |
|---|---|---|---|---|
| ETH+/ETH Convex LP (CryptoSwap) | 2.27% | 75% | 6.2m | 60% |
| Morpho RE7 WETH Vault | 2.12% | 12.5% | 1m | 12% |
| OETH | 2.59% | 12.5% | 1m | 0.6% |
| Blended Yield | 2.29% |
|---|
| Diversification Ratio | 40.6% |
|---|
Proposed Basket
| Collateral | 30d Yield (DeFiLlama) | Allocation Weight | Allocation Value ($) | Protocol Allocation / Collateral TVL |
|---|---|---|---|---|
| ETH+/ETH Convex LP (StableSwap) | 5.91% | 75% | 6.2m | 75% |
| Morpho RE7 WETH Vault | 2.12% | 12.5% | 1m | 12% |
| OETH | 2.59% | 12.5% | 1m | 0.6% |
| Blended Yield | 5.02% |
|---|
| Diversification Ratio | 40.6% |
|---|
It is worth noting that if this proposal is successful we will see ~$6.2m migrate out of the CryptoSwap and into the StableSwap Pool. While this will certainly dilute the yields in the new pool the subsequent migration of incentives by ABC Labs will mitigate some the dilution effect. I expect yield to range between 4% and 5%. However, volatile market conditions will affect this estimate.
Additional Gas Optimisations
The new plugin also significantly reduces gas costs for dgnETH actions by reducing the frequency of underlying ETH+ refreshes. Previously dgnETH would refresh every inner collateral within ETH+ on every dgnETH action; this plugin does only refreshes once every 24h, reducing gas fees significantly (~33%). Auditors have agreed that this change is low-risk since ETH+ is so much larger than dgnETH and receives so many natural interactions daily anyway.
The dollar amount being saved per transaction is difficult to quantify as it depends on gas prices but some point estimates are:
- 1 gwei: $2.70 saved per action
- 10 gwei: $27 saved per action
- 30 gwei: $81 saved per action
Risks
The difference in risk profiles between these two collateral plugins are minimal given their similarities. However, any collateral basket change carries a one-off operational risk which could lead to slippage above the value held in the backing during the migration. If this was to be the case RSR would be seized and sold off to re-collateralize the basket. This one-off operational risk will be minimised by ABC Labs careful stewardship of the basket rebalance and its accompanying auctions.
The gas optimisations that also come with this plugin also don’t come without risk. By refreshing ETH+ less frequently, there is a time lag before dgnETH sees the most up-to-date state of ETHplus. If problems develop within ETHplus (e.g. a inner asset depegs) dgnETH would only reflect this in the next refresh meaning a temporary mispricing could occur for up to 24 hours. In a crisis arbitrageurs could exploit dgnETH’s stale view during the delay by either minting or redeeming dgnETH against misaligned valuations. However, it is apparent that the benefits of these optimisations are minimal when balanced against the size of ETH and the fact it has so many natural interactions daily.
Conclusion
Adopting the StableSwap LP plugin not only brings dgnETH in line with ETHplus liquidity optimisation strategy but also improves the yield profile of the collateral basket and provides meaningful gas savings for holders. However, these benefits need to be weighed-up against the operation risk of the migration and the introduction of stale-data risk as ETHplus inner collaterals would be refreshed every 24 hours as opposed to at every action as it is now.
Careful oversight during the transition will be essential, but overall this update improves the yield profile of the basket, reduces costs and aligns with operation of the wider DTF ecosystem.
Poll
- Yay, I’m in favour the proposed basket
- Nay, Keep the current basket
- More information / Further discussion needed