Max Trade Slippage Proposal

Summary

Change Max Trade Slippage from 1% to default value of .5%.

Abstract

Max Trade Slippage is the percentage value the describes the maximum deviation from oracle prices that any trade that the protocol performs can clear at. Slippage permits additional price movement beyond the worst-case oracle price. 1% slippage means if oracles quote the price for USDT at $1.00 per USDT, the protocol will allow an auction to clear at $.99 per USDT, in the current state. I am proposing to half this possible discrepancy, to $.995 per USDT.

Problem statement

Clearing a trade of an asset at 1% deviation from the oracle price allows for those that participate in the auction to, in the limit, extract 1% of the total collateral value being auctioned from the RToken stakers. $3.75 million of USDT (if eUSD is at 7.5 million circulating supply), means that the auction participants can extract $37,500 from the RSR stakers (this is assuming there is no competition).

Rationale

I would like to change the Max Trade Slippage to the default value of .5%. 1% is too great of a value for upcoming eUSD collateral auction.

Risks

It is the default value, proposed by the Reserve team. There should be no risk to this.

Vote here:

(experiencing technical issues submitting proposals, standby)

4 Likes

Great proposal! I think 0.5% would be a great maxTradeSlippage :+1:

3 Likes

Here is the link to vote on this proposal in Register:

1 Like

Something I worry about with this proposal is that during defaults, prices may be moving quickly, so auctions could fail to clear over and over if oracle prices are changing. For this reason, I lean toward keeping the max slippage higher and relying on developing healthy competitive participation in auctions in order to reduce the slippage through competition. I don’t feel super strongly about this, but that’s my inclination. I’m going to vote against this proposal this time, but I remain open to debate and might very well be convinced it’s the right approach for a later proposal.

1 Like

Sound logic. I think Henry’s suggestion to split this variable into two different categories is interesting. Is that possible?

“I’m also curious if it is possible and might make sense to improve the backingManager at some point to have two max slippage settings so that we could have a looser max slippage setting on defaults than the one used for governance proposed basket changes.”

2 Likes