Summary
This proposal proposes to increase the backing buffer from 0.1% to 0.15% in order to reduce the risk of slashing from future collateral basket changes.
Abstract
This proposal recommends increasing the Backing Buffer from 0.1% to 0.15% to better protect RToken stakers from slashing events during basket changes. The Backing Buffer is essential in safeguarding RToken holders by maintaining extra collateral to handle slippage during a rebalance. An optimal buffer level prevents RSR slashing and stabilizes yields for RToken stakeholders. This proposal addresses a recent significant slashing event linked to insufficient backing and collateral mispricing. Raising the buffer to 0.15% aims to mitigate future risks while monitoring it’s impacts.
The Backing Buffer
The Backing Buffer is an important RToken parameter which is designed to protect RToken stakers from slashing events during basket rebalances. The percentage value describes how much extra collateral is held outside of the RToken backing relative to the market cap.
The capital buffer is then used to absorb slippage associated with collateral basket rebalances, if not present the backing would shoulder this loss trigger an RSR sell-off to re-collateralise the basket ensuring RToken holders don’t incur a loss.
Setting the backing buffer parameter requires careful balance;
- If the backing buffer is too low RSR will be seized on every basket rebalance, potentially wiping out months of yield.
- If the backing buffer is too high RToken holders and stakers yields will become sensitive to supply changes since new issuance creates a hole that must be backfilled before revenue distributions can resume and new redemptions will trigger auctions selling off a portion of the capital held in the backing buffer temporarily inflating yield.
It’s important to remember that a one-size-fits-all approach cannot be implemented for backing buffer parameters since it’s a delicate balance for multiple factors including; (i) the expected loss due to slippage of the most illiquid backing collateral, (ii) the yield generation potential of the collateral basket and (iii) the sensitivity of RToken holders to variance in yield and slashing events.
More information about the Backing Buffer and other RToken parameters can be found in the Reserve protocol documentation here.
Problem Statement
Recently the dgnETH collateral basket was rebalanced, resulting in the removal of pxETH and a 75/25 weighting of the Convex ETH+/ETH LP and Morpho Re7 WETH Vault. During this rebalance the degree of slippage was significant leading to complete depletion of the Backing Buffer and a heavy slash to RSR stakers.
While a mispricing of the Convex LP token was mainly to blame, which will be mitigated in the future with auctions being integrated into CowSwap and higher participation in auctions, the slash to RSR stakers could have been mitigated by a deeper Backing Buffer.
Rationale
Given the recent slashing event during the last collateral basket change it is prudent to consider an increase to mitigate against future RSR slashing events.
Keeping in mind the effects of an under provisioned collateral basket, this proposal suggests a 50% increase in the Backing Buffer increasing it from 0.1% to 0.15%. With this parameter constantly being reviewed after every collateral basket change to ensure it is optimally provisioned, not too high to cause severe fluctuations in yield and not too low that stakers undergo a slashing event on every rebalance.
Yield Fluctuations
Under normal circumstances an increase to the Backing Buffer means RToken holders and Stakers alike would suffer a short term loss in revenue while the buffer fills, since all yield generated by the collateral basket is routed there until it’s filled.
However with a recent reduction in the dgnETH market cap the backing buffer is currently sitting at 0.12% of the market cap meaning the drop in yield for both RToken holders and stakers would be minimal, ~1 day of yield lost. These assurances are on the provision that no auction is run on dgnETH in the time between posting this proposal and its completing on-chain vote.
If an auction was to be run in the interim we’d revert to standard operating procedure and loss of revenue would occur while the buffer fills. Under this model the backing buffer would take ~2.4 days to fill and resume yield streaming to RToken holders and RSR stakers, as shown in the graphic below.
Risks
There is no inherent risk to RToken holders or RSR stakers capital with this proposal only the potential drop in revenue while the buffer fills.
Summary
The proposal suggests increasing the Backing Buffer from 0.1% to 0.15% to decrease RSR slashing risks during future basket changes. A dgnETH collateral rebalance recently led to major slippage and buffer depletion, heavily affecting RSR stakers. Expanding the buffer will prevent similar issues by absorbing slippage. The change should have minimal yield impact, estimated at about one day’s revenue due to the current buffer level. This adjustment seeks to better protect RToken stakeholders while balancing yield stability.
Poll
- For Backing Buffer increase
- Against Backing Buffer increase
- I require more information