HackMDRocket Pool 2.0 Tokenomics, Pt 2:
Current Research Problems
-Wander, 2022/04/24 (vvander.eth)
tags: rocket-pool research
Rocket Pool has been live for about 6 months now, and in that time, we've seen quite a few suggestions on future protocol improvements. This series focuses on areas of research for Rocket Pool tokenomics and suggestions for the future – this part is about the potential areas of improvement to which tokenomics research efforts are being directed.
NO Capacity Constraints
The most obvious limiter for Rocket Pool currently is a dearth of NOs compared to rETH demand. It seems that the demand for a safe and decentralized liquid staking token like rETH, even with an above-market fee of 15% attached, is much higher than the demand for extra rewards from new stakers. Perhaps after withdrawals are implemented and existing solo stakers can freely join Rocket Pool, this imbalance will resolve itself, but with Shanghai up to a year away, this will likely be an issue for quite a while.
There are many proposals for addressing this which are debated regularly, but it's worth pointing out that Rocket Pool already offers incentives for NOs above and beyond what the Ethereum network provides for solo stakers. To keep network growth healthy, maintain these incentives is important so we can continue attracting as high a share of new stakers as possible, though incentives are likely the least efficient way to go about this.
RPL vs ETH Yield
Speaking of NO incentives, the community has also discussed the issue of ETH vs RPL yield for a long time, even pre-launch, as it's clear that RPL yield must stay above ETH yield for NOs to be properly incentivized to provide security above the minimum 10% RPL. Without that higher yield, average collateral will trend towards 10%, which is the minimum necessary to continue receiving commission. With the current protocol design, though, when the protocol has enough RPL staked, combined with higher yields for post-merge ETH, the yield on ETH may exceed that of RPL eventually. When this happens is hard to estimate right now, but most agree that it will happen eventually.
The Rocket Pool community is generally split between 1) those who believe the network as a whole will transition to a lower collateral ratio gradually enough so that it's harmless and 2) those who want to change the protocol to keep RPL yield higher than ETH yield. For the second suggestion, there are two general ways to approach this. The first is via more protocol incentives like the current NO RPL rewards, and another, which I prefer, is direct value generation into the RPL token so that it gains on the ETH ratio. Stay tuned for more discussion of this subject in a future installment.
rETH Liquidity
On the other side of the Rocket Pool system, rETH demand is extremely high. The deposit pool is full nearly 100% of the time. Several factors influence this, such as Tetranode's veCRV vote donations for ycrvwstrETH, Alchemix's new rETH collateral option, and at least one bot which appears to be taking any free space in the pool as it appears. The attractiveness of rETH as a DeFi lego is obvious and should be celebrated as a great achievement for the protocol and community.
There is a lot of room for improvement with rETH market depth, though. For integration with DeFi's biggest dapps like Aave and Maker, rETH needs to have a sizable market depth to protect those protocols in case of liquidation events. Maker has suggested a short-term goal of ~10k ETH worth of rETH market depth with a long-term slippage goal of less than 1%.
This problem is complex and somewhat circular in nature. An rETH/ETH LP is only attractive if it generates higher yield than holding pure rETH. For trading fees to outpace pure rETH gains, a high level of market trading activity must be present. Ultimately, it's unlikely that a high enough level of trading activity on secondary markets can be achieved naturally, and if so, it would likely require rETH to be ubiquitous in DeFi already, leading us right back where we started. The game theory for all the different protocol and market states here is complex, so I won't go into deeper detail here. If this interests you, please stop by the #rETH-Liquidity-Incentives thread on the RP Discord server to discuss futher.
Is there no way to stop this circular disincentivization of LPs? There is a way, but it requires doing something strange: providing liquidity incentives for a token which is already experiencing high demand. Ideally, these incentives would be paid directly with protocol-owned RPL to minimize sell pressure, but it's inevitable to see some increase in RPL sell-pressure from any rETH liquidity incentives. Is there a way to resolve this problem without devaluing RPL? Fundamentally, no. But perhaps there is a way to do so by leveraging future RPL growth instead, something I'll cover in a future installment. To understand this, though, first we need to dive into the supply dynamics of RPL.
RPL Inflation: Devaluation or Supply Crunch?
Currently, RPL inflation is 5% annually, with 70% of this allocated to RPL rewards for NOs, 15% allocated to the oDAO, and 15% allocated to the pDAO.
Among protocols with inflationary tokenomics, this is relatively tame, but it is unbounded, which typically translates to devaluation. Pocket, a protocol with a similar tokenomics design to Rocket Pool but with much higher inflation, has seen its token value decrease in relation to ETH due to inflationary sell pressure. The same has not been true so far for RPL, however, and I believe it's likely that RPL will actually increase on the ratio over time unless conditions change. In fact, a fixed 5% inflation may not be high enough based on current protocol growth. That is, protocol growth has been so strong that it's outpaced the devaluing effects of inflation. Let's do some adversarially conservative calculations to gain more clarity on this:
-
For sell pressure, assume all 75k RPL monthly inflation is sold for ETH on the market immediately
That's initial 18m RPL supply * 0.05 / 12 = 75k.
-
For buy pressure, assume 150k new RPL staked per month is all bought on the market.
This is less than the amount staked between March 20 and April 20 of this year, chosen intentionally as the lowest monthly growth rate since launch.
In this simplistic analysis, we can see that the buy pressure for RPL exceeds the sell pressure by a factor of 2x. A more fair analysis using realistic RPL restake rates (instead of selling 100% of rewards) and average growth data for the protocol (instead of the slowest month) would show buy pressure heavily outpacing sell pressure many times over. Note that 15% of inflation isn't being used currently, either, as the pDAO doesn't exist yet and the RPL treasury is therefore unable to be spent.
Growth vs Inflation
Obviously, Rocket Pool's tokenomics are not inflationary enough to offset even the least growth the protocol has seen so far, but what happens if that growth stalls further? If we again use the simplistic analysis from above, the growth rate of new RPL staked could be at least halved before inflation began to eat at the value of the token. Although this seems unlikely in the short-term, at some point the growth of the network will stall, and at this point the endless inflationary mechanism will slowly devalue RPL and begin to disincentivize collateralization of nodes and participation by NOs in general unless RPL yield remains high enough to outpace it.
Ultimately, the inflationary mechanics of Rocket Pool, while sufficient for current network conditions, may not always be sufficient. In a future installment of this series, I'll discuss some inflation mechanics changes which could suit the network's needs better both now and in the future, such as a dynamic inflation rate.
Note on Real Market Valuation
Astute observers will note that since the protocol launch in Nov 2021, the RPL/ETH ratio has remained relatively stable. If this tokenomics analysis is true, why isn't the ratio steadily increasing with protocol growth? There are several factors at play:
- It takes time for these effects to play out, and RP has only been live for ~6 months. In the short term, macro trends are not as obvious.
- There is a stabilizing factor for RPL price which comes from the protocol in the form of overcollateralized nodes selling RPL once ratio increases free up their and they have >150%, minimum 10% collateral, etc)
- A robust OTC market facilitated by Patricio's remaining 3m RPL.
- Speculation, which hides fundamental effects, likely plays a large part in RPL's current valuation. The broader crypto market indicates that a long-term upward trend ends in a highly speculative blow-off top, followed by a cool-off period before the cycle repeats. RPL may have already had its first blow-off ratio top at ~.012 right after launch.
We can't expect this situation to last forever, though. 3) may be the first to fail, as Patricio's RPL will either be staked via a Combo Node arrangment (see Part 3) or be exhausted via OTC buys. After that, it's only a matter of time until the stabilizing factor of 2) fails to contain buy pressure.
Inflation Conclusions
All this adds up to a startling realization: without any changes, even if the protocol grew at a minimal rate for long enough, it would eventually lead to a RPL supply crunch, as has been noted elsewhere (see Xer0's RPIT and Logris' analysis). While this sounds fantastic for investors, this is money left on the table until a nebulous future moon date, and the protocol could use development and improvement right now. Adjusting inflation mechanics as part of a broader tokenomics udpate which takes aim at all the points listed above is prudent.
Part 3
This is the first in a multi-part series. The third part discusses current proposals for changing the protocol in response to the problems discussed above. Part 1 discussed some difficulties with the common suggestion of ETH-only minipools.
Thanks to @knoshua and @Marceau for their feedback on this installment.
