In early April the reserve rates on all governance owned prize pools were set to 50% of the interest accrued (0% of the COMP accrued) this has led to $550,000 of reserves accruing in that time (see here). A few major things have changed since that time that I believe warrant revisiting reserve rates.
1). POOL distribution rates are being reduced by 50%
2). Treasury diversification has passed giving the protocol a $6 million reserve of stable-coins
3). At least temporarily, macro market conditions have changed leading to a decrease in underlying interest rates earned
Setting Reserve Rates
Overall, I’m of the opinion that we should lower reserve rates to optimize for larger prize sizes in the early stages of protocol growth. We are in a crucial period of time to establish the protocol as the leader in the no loss prize savings and having large headline prizes is important to both attract capital and build the brand. My opinion on this has changed since early April based on the fact that 1) treasury diversification has given the protocol a strong base of stablecoin reserves. 2) POOL distribution reductions means we should have less of an issue with aggressive yield farming.
Overall, I see an opportunity for us to aggressively focus on building the largest possible prizes. I believe PoolTogether can have a viral moment where we reach a tipping point where the prizes become exciting enough to draw all of DeFi and people outside of DeFi in. Lowering reserves rates makes this goal much more attainable.
Uses of Reserves
In the interest of tackling one issue at a time, I only want to briefly touch on this but I think we can also use reserves in more interesting ways to drive growth. Specifically, I think the ideas of 1) using a portion of the reserve to buy back POOL and distribute that to POOL pool depositors is good and 2) using a portion of the reserve to fund Gitcoin grants could have a very strong community and marketing impact.
Summary
Maintaining the 50% reserve rate for COMP and UNI prize pools make sense. Especially because the majority of those prizes are in the value of the COMP accrued not the interest earned.
For the Dai and USDC prize pools, I’d advocate something in the 0-15% range to optimize for the largest prizes.
Separately I think we should consider the best ways to utilize reserves to drive growth. I only want to plant the seed for this discussion as I believe it will take a bit more work and time to sort that out.
I think this is a step in the right direction to increase both the weekly prize value on stable coin pools (with USDC as the focus) and increasing the # of depositors / attention in DeFi. I do think a 0-15% reserve rate is quite a big leap to take in a single change. Perhaps a planned staggered reduction rate instead?
30% reserve rate as underthesea mentioned in discord is a good middle ground for now imo. Open to hear more suggestions from others!
Edit:
With some further discussion on the community call I think a lower reserve rate than 30% can still be quite viable if we are trying to reach the goal of 1m prizes per week.
I am a big fan of the current perpetual growth model and like the idea that over time with compounding interest and a high reserve rate we will be generating huge prizes down the road and alot of that with house money. I would hope we stay in the 30-50% range to keep accumulating value for the protocol and POOL holders. I am happy with the current 50% but would be flexible if we are still bringing in good value to the protocol. I think a 0-15% reserve is far too low and I wouldn’t be in support of that change.
PoolTogether can do a great service to the crypto community as a whole. Like many others, PoolTogether was my first foray into DeFi. The prizes at the time were only a couple grand. however, I had owned bitcoin and eth for years so I was already interested in crypto. Had I not been interested in crypto, I do not think that a 1 in 100,000 chance to win only $10,000 would have pushed me to learn how to use a metamask and deposit. Yes, we have Dharma, but that still requires downloading a new app, connecting a bank account, depositing, etc. We must use Human psychology to our advantage. that prize has to be enticing to attract those who otherwise would never get into crypto at all. We can have that viral moment @Leighton was talking about where mainstream adoption explodes. Once we get users into PoolTogether they will explore the rest of the ecosystem. We have an opportunity to be the diving board into the DeFi pool. Imagine pitching PoolTogether to your friend. what will be more likely to convince them to join? A $50,000 prize or a $10,000 prize? They arent going to want to hear about how the reserve rate benefits the protocol. Most people will not care about that. The reserve really does not contribute to growing that prize much with the current TVL. increasing TVL will be exponentially more effective at increasing the prizes than the reserves in sponsorship. when the TVL grows enough then we can take a reserve that will be much more meaningful to the protocol.
It has been mentioned to drop the reserve on only the USDC pool. I think this aligns with our goal to push more value into a single pool to reach our Million dollar prize goal quicker. when new people look at the different pools they will see the USDC pool has a much higher prize. thats where they will go. it will create a positive feedback loop. the larger the prize the more depositors the larger the prize gets etc…
I would like to change only the reserve rate on the USDC pool. Make it zero percent and leave the rest of the pools as is.
I wholeheartedly agree with leveraging human psychology to the protocol’s benefit. Lowering the reserve rate to increase our weekly prize pools is a critical move we need to make. The large prizes are what draw people to PoolTogether. We don’t need to stifle our growth in the short term to grow our reserves, which like you said the general public doesn’t know about, and quite frankly, won’t care about. Thanks to the awesome VC deal the put together by the team and heralded by @Leighton, we’ve got enough stables for a long time. That was a genuinely awesome team and community effort. It shows that everyone in this community is willing to compromise when needed and adapt to new information.
Even though we voted for that 50% reserve rate with like a huge majority voting in favor, I think we can appreciate the new circumstance we’re in. We should definitely decrease the reserve rate to 0-15%. With the focus on USDC, we’ll really be able to ramp up that prize pool to headline levels. We need all the juice we can get. Having more TVL locked and a higher prize pool is what ultimately generated the protocol a lot of attention beginning in February. A bunch of influential people in the space started to show their support for the protocol and that will only increase once prizes get more sizeable.
More people now than ever know how to use MetaMask now and I think they will be truly swayed into depositing into the protocol when they see a $1,000,000 weekly prize. Heck, something like $250,000 is already attention grabbing. We should do everything to grow our weekly prizes, that’ll be our growth driver. We can absolutely sacrifice short term reserves in favor of increased interest in the protocol from the wider public.
Edit: I’m seeing some discussions in the Discord about only decreasing the reserve rate to 0% for the USDC pool and leaving it be for the rest. I think that strikes a great balance between governing a growing protocol and leveraging the drip rate changes we made to DAI and USDC. We need to go full force on that USDC pool.
The reason I supported ramping the reserve up to 50% in the first place was because I saw it as dead money from a TLV acquisition perspective. I don’t really see what has changed since to make me change my mind on that; I still think the TLV is heavily determined by the POOL reward.
This would be a much more interesting idea if we shifted the POOL reward to sponsorship. I can see this move contributing to TLV then, but not really right now. See it as counter productive. The empirical evidence suggests POOL is the determining factor when users decide to deposit.
So by having usdc as the highest POOL yield pool on top of zero percent reserve rate in that pool it will become the most attractive pool for depositors. The prizes and yield will be highest in the usdc pool. This will attract new users. Organic growth should be a key factor in our decision making process and I feel that these steps will do just that. The reserve rate does not do much to grow the prize if you run the math as @Brendan has pointed out. This is a much more aggressive approach to grow tvl. The other pools would still have a reserve rate so we would still accrue value to the protocol
My argument is that if the reserve rate is 0% or 50% it is not going to make much, if any, difference to the TLV of that pool. We have empirical evidence of this. Before pooltogether had a token the TLV was what? 1-2mm? and once POOL launches it skyrockets.
It’s not in my interest to knock organic growth; I want organic growth, but I think people are making gross assumptions that this is going to boost TLV. If it was a trigger when we ramped it up to 50% why did we get no negative reaction from the TLV? There is strong evidence to support that it will make no difference.
What about reducing the reserve rate to 30% and using it as follows:
10% Reserve
10% Distributed to POOL holders (via POOL pool staking)
10% Distributed to depositers.
The numbers can be played with but this could be an interesting experiment to see how it effects TVL. This way keeps the reserve growing at a smaller rate but also rewards both POOL token holders and depositers of the pool which could help drive the price of POOL and also further incentivise deposits as they’ll essentially recieive a small % return even if they don’t win. This could also be considered for the pools that we lower the POOL drip rate for as it can help replace some of that reward.
Well, we’ll be giving away most of the revenue we’re currently making; mainly to whales that are yield farmers and will dump their POOL tokens too. Nice unexpected bonus for them I guess.
Reserve rate has nothing to do with POOL distribution though. We would be sacrificing a small amount of reserve for now to create TVL growth. It’s a long term play. We can turn the reserve back on in the future when the pools are much larger. Then the reserve we capture will more than make up for what we miss out on short term. I think this is honestly a no brainer. It aligns with our goal to have one headline pool that attracts attention.
Lots of good discussion on this thread! I think @ageless summarized the counter argument well.
My argument is that if the reserve rate is 0% or 50% it is not going to make much, if any, difference to the TLV of that pool. We have empirical evidence of this. Before pooltogether had a token the TLV was what? 1-2mm? and once POOL launches it skyrockets.
This is great point! Looking at the data, the 50% reserve rate went into effect on April 8th. Using the USDC prize pool as a sample we actually saw a pretty dramatic INCREASE in total deposits after April 8th.
This increase in deposits, was despite a decrease in total prize sizes (dropping under 100k per week). However, there was an increase in POOL token price (leading to an increase in effective APR for depositors)
So this data seems to pretty concretely show that right now the primary draw for depositors is the effective APR and the prize size doesn’t make a radical difference.
All the above notwithstanding, I do still think there is an argument to be made that hitting top line prize numbers will accelerate growth. It’s hard to know the impact though until we actually do.
One other important piece of data, thus far, what has really drawn press coverage is not the size of the prizes but rather small depositors winning large prizes. Specifically, our “luckiest” depositor who put in $73 and won $43,000 was covered by press (here and elsewhere).
So I think based on feedback, my current summary is:
Don’t change reserve rates right now. Wait to see impact of POOL distribution changes. Additionally, if reserve rates are going to be changed it probably needs to be on the thesis that larger prizes drive marketing which drives TVL and not that larger prizes directly drive TVL
Continue to work on ways to help small depositors win (pods should be a big help here)
Continue the discussion on more interesting ways to use the proceeds from the reserve
I still think the most effective thing we can do is completely shift the POOL drip to sponsorship. Then reducing the reserve rate will become a trigger for increasing TLV.
100% this! Until the POOL drip is shifted primarily (or hopefully entirely) to the sponsorship pools, we won’t be able to get the necessary data to make informed decisions on reserve rate and incentives.
I agree with @Leighton that we should prioritize growth at this stage and that the size of prizes is likely important, particularly at inflection points like $100K or $1M/week. However, until we have a better sense on how much of the deposits are driven by actually wanting to win prizes as opposed to farming POOL, I don’t think we should meaningfully reduce the reserve rate.
