Add 0.01% fee pools to Volume Splitting Groups - OSMO/ETH, TIA/USDC, OSMO/USDC, OSMO/USDT

Proposal Discussion Liquidity Incentives
1

This proposal would create new volume-splitting groups to include 0.01% spread pools for incentivized pairings that have a 0.05% pool in dominant use.

Background

This proposal extends the Volume Splitting Groups for existing groupings to include 0.01% Spread Factor pools after promising results from adding 0.01% and 0.00% OSMO/ATOM pools in Proposal 718.

Early results show that since the lower fee pools were included on the 12th of February, there have been beneficial results for Osmosis’ incentives sustainability and trader experience with no additional incentives expended.

Lower fee pools within the group have quickly gained a large volume share, catering to over 50% of all ATOM/OSMO trading volume on-chain. As this volume has increased, more incentives have been pulled towards these pools, attracting more liquidity and enabling more volume to pass through these pools.

This, in turn, has led to volumes in this pairing crossing the trend line of volume overall in what was a declining trading volume pair due to the lowered overall fee.

image
image1806×724 130 KB

The same Osmosis incentives supplement these lower fee pools as applied previously to higher swap fee pools. However, as the Protocol Taker Fee remains constant with a changing pool spread factor, a greater proportion of trading fees go to the protocol and are either retained in the Community pool or distributed to Stakers. This ratio of fees going to the protocol has rapidly increased in the ATOM/OSMO pairing from where it had leveled off at 57% to a share of 75%.

A primary KPI for the performance of incentives is whether a protocol pays out more for liquidity than it earns in return. While Osmosis has previously crossed the Fee Subsidy rate of 1 overall, meaning that pool liquidity is mainly sustained by swap fees rather than incentives, it still pays out more than is earned by the protocol itself.
Implementing these low-fee pools and further incentive optimization initiatives reduces this towards the 1:1 target.

image
image1806×733 96.3 KB

Importantly, this has resulted in a typical swap fee paid by traders between ATOM and OSMO reducing from 0.18% to 0.14%. By expanding these low-fee pools to other major pairings, the typical swap fee on Osmosis will continue to lower, particularly for multi-hops through these pools onto pools that have not had the volume to sustain a low swap fee pool without incentives or multiple quote asset pairings.

image
image1813×738 45.8 KB

The complete statistics of the pools within this grouping, which comprise the data behind the charts above, can be found here.

Proposed Groupings

OSMO/STABLE

OSMO/ETH

TIA/STABLE

Target On-chain date: 25th February 2024

Related Posts:

2 Likes
2

Certainly in favor of this, but thinking ahead… should we not also include the 0% pool in here as well to avoid needing another proposal in the (near) future?

3

As the main proponent of the 0% pools, of these pairs I think only OSMO/ETH is suitable currently, and ideally I was waiting to see volume grow more overall before proposing it personally (TIA/OSMO has more volume but is a lot more volatile, which means both more fees to capture and more expensive to manage).

It’s not included in this proposal, but the one other 0% pool that I would deem to be worthwhile at this time would be USDC/USDT, and the one I would plan to propose next.

1 Like
4

I believe that ATOM/OSMO 0% is the only valid one right now due to the fees being more than the swap fees generated in an equivalent pool. This is mostly due to it being the traditionally largest pool and being the sole provider of the TWAP parameter in the incentives system until last month.

USDC/USDT is struggling to gain liquidity depth since the 0.01% fee with this subsidy is providing lower returns than stable pools elsewhere.

I would rather see the fees in the 0.01% fee pool increase to the point where we have a reliable 10M+ liquidity pool there before we push fees down any more. 0.03% total is already incredibly competitive as a conversion rate between USDC and USDT.

1 Like
5

But is adding them to the VSG not part of the solution?
I mean, the VSG should bring incentives to the spot where they are needed, right?

So if you have a pool with high volume, but low liquidity the VSG should put more incentives on that pool to make being a LPer in that pool more attractive, right?

6

Within a grouping, yes, but the overall incentives are still too low in this grouping to support size in a 0% pool.

Would we rather have 100k liquidity incentivized to be very efficient but unable to handle size on a major onramp pool since it is incredibly narrow, or 1m liquidity that is less efficient but able to handle a multiple of the size?

0% fees are unscalable since they can only ever reach what incentives can rent rather than scaling with volume. 0.01% is probably also that for volatile pools, which is why I am only proposing these when there is an established tier system of pools to provide that more self sustaining liquidity.

1 Like
7

Good point.

Higher volumes in a 0.01% pool will have a higher APR for the LPers as effect, whereas in the 0% pool this will not be the case.

Which brings me to the following question; suppose that the 0% pools will be hugely successful, should we even want it on the exchange? Because liquidity should also be supported by incentives? And if we want to get rid of inflation in due time, we can’t go with that principle.

So maybe we should go for a bottom accepted swap fee?

1 Like
8

This has passed and will be included in March’s Incentives proposal.

If anyone wishes to monitor how this impacts the volume, protocol revenue KPIs and typical user fees then I have created spreadsheets similar to the ATOM/OSMO one.
These aren’t the best sheets ever as they need an update triggering each day but they can give an idea of how these metrics are performing:

OSMO/STABLE liquidity is spread amongst the OSMO/USDC 0.2%, OSMO/USDT 0.2% and OSMO/USDC 0.05% pools, however the volume is dominated by OSMO/USDC, and recently the 0.05% pool of this.
This grouping has hovered around the target for Osmosis incentives, this should maintain with the new pool, even with incentives projected to rise in March. (Chaos Labs)

ETH/OSMO liquidity has been primarily in the 0.2% pools; however, the 0.05% pool has dominated volume since it gained liquidity.
This grouping has not yet been profitable for Osmosis, but the addition of a 0.01% pool along with the projected reduction to total incentives in the March adjustment (Chaos Labs) should easily push this under the target ratio.

Main takeaway is that almost all volume goes through the 0.05% TIA/USDC pool, 0.2% TIA/USDC appeared with incentives and volatility but overall this is dominated by one pool. This group has been profitable for Osmosis since the original bootstrapping incentives ended.

2 Likes
9

No these are by far the best sheets, pretty much ever. If you published how to make these I imagine you would start a bit of a fan club.

10

If you want them for another collection of pools:

  • File, make copy

  • Edit the cells Lookup!I2 and Lookup!K2 to include the pool number you want, I1 is just the label for convenience

  • Main!L2 changes the name of the label on the chart

  • Copy the formulae from the bottom line of main all the way up across all rows. I’ve been pasting these as values since the API via sheets is quite unreliable.

  • Either repeat steps 1-3 for the further pools on the lookup page, or delete all the extra pools on the main/lookup page.

  • Incentives need manually adding from previous proposals, however I’ll be assembling these sheet for all groups as we go.

These could be made way better by having either a user interface section added, having a more efficient lookup section from the API, or being made in something that wasn’t google sheets at all. But way faster for me to put these together and share in sheets :slight_smile:

1 Like
11

Really cool idea, this would go well with the basic pool_data page I’m putting together.

12

And this is how synergies are being created :slight_smile: