Custom Taker Fee on Stable to Stable Trades

To be posted to Snapshot July 16th, 2023


  • Carbon has a protocol wide fee of 20 BP (basis points).
  • This fee, while appropriate for volatile pairs - is not in line with the market when it comes to stable to stable trades.
  • For reference, Uniswap added a 1 BP fee option (0.01%) - in November 2021 (link)
  • This proposal seeks to take this one step further and introduce a fee of 0.001% on stable to stable trades. This is 1/10th of a single basis point.


  • A new function in the contracts will allow the DAO to identify stable pairs that should have a fee that overrides the default fee (20BP) on a given pair.
    • The initial list is further down in this proposal.
  • A new function in the contracts will allow anyone to query the fee on any pair.

Supporting Data

  • The current fee on stable to stable pairs in 20x that of uniswap
  • There are multiple stable to stable pairs on Carbon and they are not being interacted with frequently or at all.

Initial List of Stable Pairs to be Updated with 1/10th BP fee


*Note: In order to be included in the list above, a stable pair needs to exist on Carbon. If I missed any, or you would like to add one (and the pair exists on Carbon), tell me in the comments. As time goes by, the Bancor DAO can add more stable to stable pairs to this list (the contracts cannot autonomously determine which tokens are stable).


  • Gas for takers on these pairs will not change
  • Gas for takers on all other pairs will rise by ~2300 (a couple cents at today’s rates)

FOR: Update the taker fee to 1/10th of a single basis point for the stable pairs listed above

AGAINST: Take no Action


Just noted the brilliance here. By opening the door to efficiently arbing stables at a lower cost than the industry, you open the door to pulling massive liquidity into the protocol where the activation energy needed to try other strategies is as low as it can get. The fee generation isn’t necessarily in the stable strategies, but as a second order benefit to the protocol to onboard users with the initial funding for strategy creation and testing user experience.

It’s like a free customer acquisition strategy since the fees were generally cost prohibitive for stables vs. the industry.

I’m mad I didn’t think of that first.

1 Like

We fully support this proposal. @PaperStreetCapital described well what are the benefits and second order effects of this so no further comment on our side.


Okay, just so I’m sure I understand the effect of the proposal and its intent; Carbon effectively surrenders ever making any viable fee revenue from stable:stable trades, thereby slowing the process of deficit remediation. The HOPE is that stable:stable traders will then use Carbon to create other strategies which generate fees. My concern is that the kind of people who arb stables and would be attracted to use Carbon purely because of lower fees (therefore higher margins) are exactly the same who are primarily motivated by lower fees, and thus won’t be inclined to use it for nonstable trades if the fees there are still at current levels or in line with the rest of the industry.

Maybe reduce the fee to slightly below Uniswap to enable arbitrage, but not down to 1/10th and just wipe out all fee generation from stable:stable. That doesn’t sound like “a step further”, it sounds like a whole order of magnitude, which may not be necessary.

If Uniswap are operating on 1BP, why not start Carbon off at 0.8BP, instead of just completely giving up on fees which were the primary mechanism outlined for deficit reduction when Carbon was introduced? A 20% edge vs Uni still seems pretty generous, and assuming that stable:stable traders heavily automate arbitrage and thrive on small edges, could still generate 8x as many fees towards deficit reduction as just jumping straight towards the proposed 0.1BP would.

It feels bad backing any proposal whose only guaranteed (rather than hoped or speculated) result or consequence would be massive reduction in fee generation.