Summary
Rubicon’s Taker Fee will be lowered from 0.04% to 0.01%.
Motivation
The main motivating factor for this upgrade is to enable more trading volume, specifically on stablecoin/stablecoin pairs.
Most stablecoin volumes on L2 networks are on Uniswap pools with a 0.01% fee. Even when the Rubicon order books have a lower spread on a stablecoin pair, having a fee four times higher than the alternative venue poses a major friction to traders and arbitrageurs.
Implementation
The setMakerFee() function will be called on the Arbitrum, Base, and Optimism deployments of the RubiconMarket contract. The new makerFee value will be 8, which with the protocolFee of 2, means the total Taker Fee paid by traders will be 0.01%.
Considerations
This upgrade decreases the maker rebate. However, Rubicon market makers are already earning additional rewards that significantly outweigh rebates, and will also directly benefit from the increase in trading volumes.
1 Like
Thank you @forrest for getting this started. I wanted to weigh in on some of the motivations for this change from my perspective.
The balancing act with this decision IMO is maker rebate vs how competitive our market makers can be (fill rate). With a fee rate of 0.04%, we noticed something really interesting from market data: it is hard for Rubicon market makers to offer the best network rate on stablecoin pairs. With many AMMs using a 1bp (0.01%) fee for stablecoin pairs, it becomes infeasible for Rubicon market makers to reliably outbid the AMM. That is a really important feature of our system, offering the best rate on the network, and this is the main driving force for the change. In short, by reducing the fee, we expect Rubicon market makers can do multiples on their competitiveness and fill rates (earn more spreads and rebates) by being able to offer tighter spreads net of execution costs for Rubicon takers.
An essential concept to the competitive dynamics of Rubicon vs other venues and a core reason why Rubicon is great is that market makers can offer the best rate on any given network, down to the wei. The active Rubicon market maker can outbid static AMMs, block to block, and offer the tightest spread - we call this NBBO (Network Best Bid Offer). When offering NBBO, market makers on Rubicon get reliable demand and fill from cost-motivated takers and arbitrageurs on the network. By adjusting our fee rate down to a total fee of 1bp with this change, it becomes much easier for our market makers to do this on all pairs. Through a different lens, in microeconomics P * V = Rev for the market maker, and we believe by lowering the price, as is the classical relationship in this function, that the gains in volume our market makers can earn (as well as the competitiveness of their quoting) will outweigh the reduction in price (rebate reduction) leading to a net increase in revenue.
Although it doesn’t feel great lowering the maker rebate, we feel that the benefits from the change, namely an increase in fill and competitiveness for Rubicon market makers, will outweigh the rebate reduction.
Please feel free to weigh in on this and let us know your opinions. We aim to iterate and improve every day, learning from market data in an unbiased way to make the best decisions possible for the protocol. One day a decision like this will be properly enacted via DAO processes - until that day we aim to communicate our thinking and decision-making as clearly as possible and welcome all criticism and insight.
Ad Victoriam,
Benjamin
2 Likes