This is a cool idea I hadn’t thought of! But I think that this violates the mandates of proposal 629 with respect to what the council’s actual role is.
There’s nothing stopping the council from making adjustments after this default fee is turned on though to conduct such a study is there? On the contrary, it seems like the council would also have the benefit of additional data by doing it this way.
e.g., let’s say you have pool A and pool B, each having different assets and pairs. You could turn them on at different fee rates from the start, but since the assets are different any variance in volume decreases or TVL decreases could be attributable to other factors beyond just the fee itself.
If you turned them on at the same rate, waited a week or two, and then adjusted, you would have the benefit of seeing the impact of multiple fee levels, which is just an additional data point you wouldn’t have had before, right?
@floppybanana This seems more like something the subdao should be working on. I have thoughts here but gov has agreed to delegate this role to the subdao (And frankly I don’t have a lot of time myself to sketch out a spec for this)
Initial thoughts though:
As you wisely pointed out above, there’s a rate at which the reduction in volume begins to have a disproportionate impact on the revenues collected by the taker fee. Think we can all agree for example that a 10% taker fee would destroy volume such that it defeats the purpose of having a taker fee in the first place
Once fees are turned on and we have volume reduction data points to look at, I’d love to see someone map out a curve reflecting anticipated / estimated volume reduction levels at multiple fee points based on those existing data points. This could be used by governance to make adjustments as needed and find the “sweet spot” at which fee revenues are maximized and volume impacts are minimized, taking into account other factors such as impact on pool fee APRs (which impacts organic growth of liquidity) and impact on strategically valuable pairs.
However, as mentioned above I feel that there’s no reasonable way to find this “sweet spot” of fees until we actually turn them on. 0.1% seems like a reasonable level at which to do so in order to get a good sense of impact without causing significant damage to the protocol’s volume.
As said, I do agree with the standpoint @whitemarlin made that it already has passed governance, but I still remain wary that a lot of people only noted the potential money to be made without thinking of the trade-offs (which I have seen happening to often already in governance…)
In this sub-DAO we have multiple parties that have mentioned their support for an initial 0.05% rollout as plausible. This post also shares a twitter poll suggesting 0.05% as the preferred choice. Mu hunch here is that rolling with 0.1% vote on-chain is going to see controversial outcomes.
We were supposed to have a DAO vote to define if we supported a reduced initial rollout before building a mathematical model to guide our fee adjustments over time. OR if we decided to build the model first and rollout full force after.
I would understand a vote to go on-chain and let the whole community decide on this matter. But here we have a totally different story. This vote recommends an initial rollout of 0.1% which double the amount the DAO was considering in our chats.
This seems quite problematic to me. I’d advise delaying the decision to let the DAO settle its own internal vote first OR at least lowering the amount to the 0.05% most people tend to agree on.
In this sub-DAO we have multiple parties that have mentioned their support for an initial 0.05% rollout as plausible. This post also shares a twitter poll suggesting 0.05% as the preferred choice. Mu hunch here is that rolling with 0.1% vote on-chain is going to see controversial outcomes.
I think you misunderstand the scope of the subdao. Community governance is responsible for setting the default rate and cannot proceed until community governance sets the default rate. The subdao scope is to set the non-default rate for pools that need more specific rates. At all times, community governance holds veto power of what the subdao sets. As Robo noted above, based on the language of the Subdao prop, the subdao cannot proceed until community governance sets the default rate.
Community governance has taken proper precautions such as waiting for the lowering of pool fees from 0.3% to 0.2%, etc.
In summary… 2 things going on here.
Default rate - This is the rate set by community governance. It is currently set to 0.00% and this prop would turn it on at 0.1%
Non-Default rates - This is the responsibility of the subdao.
After reviewing the information, I acknowledge that you are right. The DAO should only take care of recommendations for the alternate fee levels as you mentioned.
Still, regarding this proposition, I think many parties (including some in the DAO) have showed support for a lower 0.05% fee rollout as an easier starting point to assess the dynamics this new fee will create. I was actually saying that maybe this proposal could take notice of the different “recommendations” made by some members of the DAO. But you seem to imply that we are overstepping our mandate and I understand that decision.
Therefore on a personal level, aside from our role at the DAO, I’d like to say that PRO delegators doesn’t support a 0.1% initial rollout for the fee. We think a 0.05% test for a month is more cautious and reasonable in terms of risk management. If nothing out of the ordinary occurs for at least a month then we will be supportive for an increase to the 0.1%.
Still, regarding our position in the DAO, we will respect governance and align with the majority vote if the community decides to start with 0.1%, we will then work with the other DAO members on our recommendations based on this default fee.
Another plus; a slighltly lower starting point at 0.05% gives the opportunity to build the required models in a scenario where the taker fee is applied, but still at a low level. That way we can test the model and see what happens before we go full force.
Based on sentiment on twitter and previous sentiment expressed by community will put the proposal up to set it up at 0.1%. If the community feels this is too high we can vote “no” and revote on 0.05%.
0.1% represents a $1 fee on $1,000 transaction. Combined with spread factor of 0.2% it is competitive with other exchanges at 0.3%.
0.1% also provides for more flexibility as it can be adjusted up or down 0.05% by community based on the results.
Based on feedback in this thread I gave an extra 11 days for the subdao to establish framework for how to make exceptions to the rule set forth by governance. During this time, the subdao has voted on and passed a framework 5 days ago.
Is the subDAO planning on making an immediate exception for the USDC.AXL/USDC transmutter pool or will the 0.1% fee be applied to the pool as well?
Is the subDAO going to be sharing its analysis framework with the community?
Can the subDAO members use their names on the DAODAO DAO so the community members know who voted which way and who actually participated in votes. I noticed 2 of the 7 members didn’t vote.
I was one of them, completely missed the time window. No excuses possible.
I am ok with the framework though, whereas the following steps need to be taken first (and soon), especially step 1 and step 2 to create the list of pools which need reduced taker fees and to bring that to governance.
step 1: Create a simplified model to define the most sensitive pairings of denoms that need to be assigned to the “zero” fee list.
step 2: Create a simplified model to define which pairings of denoms should be assigned a reduced fee (20% of the default fee).
step 3: Define an initial observation period to readjust after the rollout phase.
step 4: Assign tasks to members in order to create a more detailed model.
So Osmosis is looking to complete the first full week of taker fee implementation at around $20m volume which is approximately half of the recent week’s averages of $40m+ (worst in recent memory being $27m – I understand the week’s not complete yet but I don’t think there’s going to be $20m volume today and tomorrow, I could be wrong). Does the sub dao have any statistics to share on whether this data is providing the wrong outlook?
It is likely to continue this way, or get worse going forward in the short term. We had 1-2 great days as part of this week’s volume calc. As a group of us have been saying, the prior projections expecting the drop in volume to be modest were never realistic. We have a higher fee structure in many cases now, and more complicated to understand.
Also worth noting that we have had bad weeks end of July for example as well.
It is quite hard to assess, judge and steer on 1 week outcome, trends are way better indicators.
Sure, happy to see what happens. Just mentioning to see if there’s any interesting data collected by the sub DAO or if it’s working off of data like the one on osmosis.zone. I guess I would assume volume should trend up in a vacuum as well, since the front-end was broken and a lot of people were unable to swap. Also, it looks like multihop discount was not actually removed until today with the v20 upgrade, so we’ll see the effects of that in the coming weeks as well (although due to incentive removal, the only affected pools are supercharged ATOM, ETH, BTC – granted these are some of the largest pools though).
