Out of curiosity -- how much margin do you think professional MMs make on an average trade, in bps?
Conversation
I would assume some hilariously tiny amount! The point isn’t the margin per trade, it’s the return on invested capital
5
19
> it’s the return on invested capital
Got some bad news for you about AMMs... :D
1
That’s exactly the point Hayden is making!
Professional market makers make much higher returns on their capital than liquidity providers make on Uniswap.
Their margin is our opportunity
1
3
I don't think this is the right framing.
Are you implying that AMM LPs are willing to accept a % fee that is so low that sophisticated MMs (and other competing AMMs) just won't bother participating in the market?
Quote Tweet
Replying to @srndptme and @cyounessi1
The following image demonstrates how sophisticated MMs can eat into AMM fees until it is no longer viable to overcome impermanent loss.
2
1
6
No—my point is that AMMs can deploy way more capital because LPs are OK with modest returns and are guaranteed their fair share of fills.
Retail investors, too, are better off trading against an AMM, because liquidity doesn’t tactically dry up.
4
6
I agree that professional market makers undercutting the spread do ruin this nice equilibrium!
The solution to this, as for so many other problems, is to have all trading in the world go through Uniswap
2
9
you should see what happens when a country implements 30bp fees.
(basically retail ends up gambling on options and everyone else trades futures since spot is too expensive.)
2
1
I’m curious why you think 30bp is the only fee possible in AMM?
1
2
AMM providers get destroyed without large maker rebates because they're forced to provide at the market, both directions, whatever that is.
2
3
even with 30bp taker fees/maker rebates there's significant money to be made each day trading against AMMs, just doing arbs.
If fees were lower, AMMs would lose even more to that.
Arbitrage profits are not equal to liquidity provider losses!
Ask Kyle to explain volatility harvesting to you
1
9
They are unless the person providing in the AMM strongly things that on long timescales the assets are mean-reverty
which I think is not really what people are trying to express
and if they are then AMMs make more sense
as long as the belief is both strong and very nonspecific
3
2
Show replies
The fees may be charged in the future at a varying rate depending on such factors like market volatility. In the end I believe there’s a way to make sure LPs make a clean fee even in an “almost surely” sense since they provide the liquidity. By clean I mean no IL whatsoever
Thoughts on slip-based fees for better compensation to LP’s? Some of you may be familiar with this model:
1
Yeah, it doesn’t work (at least without other safeguards)—you can just split your trade into arbitrarily small pieces to pay arbitrarily low fees
2
Show replies