The Case for LSD
Liquid Staking Derivatives
Liquid Staking Derivatives (“LSD”) are one of the biggest themes in crypto. This past week’s successful Shapella upgrade made Ethereum staking, and a result LSD, even more prominent. The ability to now unstake ETH is the catalyst for a flood of more ETH to be staked. The upgrade is consequential for Ethereum and ever more so for LSDs.
The investment thesis for LSDs is:
LSD is the best way to stake.
The amount of staked ETH is going to increase by 4x.
Rocket Pool will benefit from the increase in staked ETH.
Shapella is a catalyst.
I elaborate on the investment case in this article, outline Rocket Pool valuation and its two key value drivers.
The investment case for LSD:
1. LSD services are the best way to stake.
There are two reasons to stake ETH. To generate an income. To ensure the security of the Ethereum blockchain.
There are three ways to stake ETH.
Solo staking: As the title suggests, it’s an individual staking ETH. It requires running a node. Operating a node requires running and maintaining an execution and consensus layer client. Solo stakers ‘stake’ (ie. deposit) 32 ETH to run a node.
Staking-as-a-service: These are professional node operators. They optimize their systems and hardware to generate the most income. ETH holders deposit their ETH with these service providers who stake it on their behalf in exchange for a fee. Cloud, Blockdeamon and Figment are examples of staking-as-a-service.
Liquid staking derivates (“LSD”) applications: LSD applications allow users to earn staking rewards with improved liquidity. Like staking-as-a-service, LSD applications stake ETH on behalf of depositors. Unlike staking-as-a-service and solo staking, ETH staked via LSDs is liquid. Typically, when ETH is staked, it is locked up until it is unstaked. LSD applications use smart contracts to create a liquid tokenized representation of the ETH staked. The tokenized representation are called liquid staking derivatives. They are tokenized claims, or an IOU, on the staked ETH.
Staking-as-a-service and LSD applications are more accessible to the average user than solo staking.
In addition to being more user friendly, LSD has two major benefits:
Lower capital requirements: Staking ETH requires a minimum of 32 ETH, worth $64,000 (at $2,000 ETH). ETH holders can stake via an LSD service with as little as 0.1 ETH. LSD services pool ETH from depositors. The pooled ETH is used by LSD node operators to stake. Node operators stake your ETH on your behalf in exchange for a fee. It means that anyone can participate in staking without accumulating a war chest worth of ETH.
Unlocks staked ETH: The drawback to staking ETH is locking up capital. Once your ETH is staked, there is nothing else you can do with it. There is an opportunity cost to staking ETH. If you stake it, you can’t use it to collateralize a loan for example. That’s where the ‘liquid derivative’ part comes in. LSD applications mint an ETH derivative for each ETH that is staked through its service. The ETH derivative can be used in DeFi applications. The ETH derivative can be exchanged for other digital assets or used to collateralize a loan. The ETH derivative reduces the opportunity cost of staking ETH. If you can do nearly the same thing with the ETH derivative as you can with ETH itself, you might as well stake your ETH and earn income.
These benefits are the reason why most ETH will be staked via LSD applications. LSD accounted for 5% of the staking market in early 2021 when ETH staking launched on the Beacon Chain. LSD represents 33% of all ETH staked today. Centralized exchanges (“CEX”), the likes of Coinbase, Binance and Kraken, represent 28% of ETH staked. Unidentified stakers, which are mostly solo stakers, have 22% share. Staking pools are mostly staking-as-a-service products, they represent 17% of the staked ETH market.
Lido is the largest ETH staker and preeminent LSD application. It has a dominant 32% market share of all ETH staked. Rocket Pool is the second largest LSD application and has 2.5% share.
2. The amount of staked ETH is going to increase by 4x
There are 18 million ETH tokens staked, which is 15% of total ETH supply. It pales in comparison to other Proof of Stake chains. The table below benchmarks PoS chains with a market cap over $1 billion. Ethereum is an outlier. It is by far the largest by market cap and offers one of the best real yields, yet the % staked is minuscule by comparison.
There are two structural reasons why Ethereum’s % staked is small.
PoS is new for Ethereum: Ethereum converted from PoW to PoS on September 15, 2022. The staking that was done prior to PoS was a form of testing on Ethereum’s Beacon Chain. PoS, and its required staking, is now mainstream on Ethereum.
Staking lockup: Eth staked is locked up indefinitely. Understandably, ETH holders are reluctant to stake their ETH if they don’t know when they can get it back.
The Shapella upgrade that went live April 12 allows ETH to be unstaked. The combination of Ethereum being a fully functioning PoS chain and staked ETH no longer being locked up indefinitely will precipitate a flood of ETH staking.
The amount of ETH staked will increase by 3.5x if ETH achieves a 50% staked ratio. Ethereum’s PoS peers have an average of 54% staked ratio. LSD’s could enable Ethereum’s staked ratio to be considerably higher than its PoS peers. LSD’s reduce the opportunity cost of staking. There is no economic reason why not to stake. The amount of ETH staked would grow by nearly 5x if Ethereum staked ratio reached 70%.
3. Rocket Pool will benefit from increased ETH staking
Rocket Pool will benefit from increased in ETH staking for three reasons:
i) Small base
Rocket Pool is less than 1/10th the size of Lido. It’s more levered than Lido to increase staking. For example, if Rocket Pool’s share increases from 2.5% to 12.5%, that’s a 417% increase in staked ETH via Rocket Pool. If Lido benefits from a similar 10% increase in share, that’s only a 32% increase. The growth in the amount staked is what matters. It’s what drives the LSD token price.
ii) Decentralization:
The Ethereum community prioritizes decentralization. I believe that the marginal node operator will prefer to support a staking application that promotes decentralization.
Rocket Pool is more decentralized than Lido for two reasons:
Permissionless: Anyone can be a Rocket Pool node operator. Rocket Pool node operators are incentivized to be good actors. They stake their own ETH and need to buy RPL and post RPL tokens as collateral. Their ETH and their RPL tokens are slashed if they act maliciously. Lido is permissioned. Lido node operators are vetted and approved by the Lido DAO. Lido node operators do not need to stake their own ETH nor do they post any form of collateral. Lido node operators face no economic penalty for acting maliciously. A malicious node risks losing ETH staked by Lido depositors. Therefore, the threshold to become a node operator is high. That’s why there are only 29 Lido node operators.
Node operators: Rocket Pool has 2,200 node operators compared to Lido’s 29. The 5.6 million of ETH staked on Lido, which represents 32% of all ETH staked, is done by only 29 operators.
iii) Node operator unit economics
A node operator is better off staking using Rocket Pool than solo staking. Rocket Pool node operators benefit from the Rocket Pool incentive fee and RPL token issuance in addition to the ETH staking income they would earn as a solo staker.
Using Rocket Pool is more capital efficient for a node operator than solo staking. A solo staker needs to stake 32 ETH, worth $64,000. A node operator using Rocket Pool requires 16 ETH. The remaining 16 ETH, for a total of 32 ETH required to be an Ethereum validator, is contributed by Rocket Pool depositors. Rocket Pool will decrease the minimum ETH required to 8 ETH from 16 ETH in the coming week(s).
Rocket Pool node operators need to buy and post RPL tokens as collateral. They need to post as collateral the equivalent of 10-150% of ETH staked in RPL tokens. Today, for example, a Rocket Pool operator stakes 16 ETH and must buy 1.6-24 ETH worth of RPL to be held as collateral.
The mandatory RPL token purchase by Rocket Pool node operators serves three purposes:
Insurance: If the node operator is a bad actor, they lose the ETH they staked and the RPL tied up as collateral. There is a big incentive to not act maliciously.
Pro-rata token distribution: New RPL tokens distributed to node operators are issued based on the pro-rata ownership of RPL tokens.
Constant demand for RPL token: The RPL collateral requirement creates a constant buying demand for RPL token.
Rocket Pool node operators generate 0.5-1.9% higher return staking through Rocket Pool than solo staking. Lido is the most capital efficient LSD application. Lido node operators do not need to stake any ETH nor post any collateral. Lido takes 10% of the fees generated from the ETH staked on behalf of Lido depositors. Lido node operators get 50% of the 10% fee.
The table below illustrates the unit economics for running a node as a solo staker and Rocket Pool and Lido node operator.
Solo stakers are better off staking via Rocket Pool.
The table above assumes Rocket Pool node operators deposit 10% of the eth staked as collateral. The average across node operators is 73%. Tying up 63% more capital yields lower annual returns for the Rocket Pool node operator, assuming a constant RPL token price. There is a much higher upfront investment.
Node operators who buy more than the minimum 10% collateral requirement believe the RPL token price will increase. The increase in RPL token price is the largest driver of their returns. The driver of returns for the node operator who deposits 10% collateral are Rocket Pool incentive fees and ETH staking income.
4. Shapella upgrade is a catalyst
The Shapella upgrade introduces two catalysts.
i) Flood of staking
Now that staked ETH can be unstaked more people will stake their ETH to generate an additional return. Gone are the days of tying up ETH indefinitely.
ii) Solo stakers switch to Rocket Pool
I believe solo stakers will choose to switch to staking via Rocket Pool instead of continuing to stake by themselves because:
Improved unit economics - Rocket Pool stakers make a higher return than solo stakers.
Stake more ETH - Solo stakers can stake 2-4x more ETH when staking via Rocket Pool. A solo staker stakes 32 ETH per node. Using Rocket Pool, that same 32 ETH can be used to stake a total of 64-128 ETH. For example, the 32 ETH the solo staker contributes is split into 2 pools with 16 ETH each. 16 additional ETH are contributed to each pool by Rocket Pool depositors for a total of 32 eth in each of the 2 pools. Rocket Pool’s imminent Atlas upgrade will reduce the amount of ETH a node operator needs to contribute from 16 to 8 ETH. After the upgrade, the 32 ETH contributed by a solo staker could be used to stake 128 ETH.
Easy switch from solo staking to Rocket Pool staking - Rocket Pool’s Atlas upgrade enables solo stakers to switch their staked ETH directly into Rocket Pool. Solo stakers do not need to withdraw their ETH, which takes time and can incur a tax liability.
Decentralized - Solo stakers embody the Ethereum ethos of decentralization. I suspect they’ll prefer to stake via a service that espouses decentralization.
The potential solo staking conversion to LSD could be enormous for Rocket Pool. There is 4 million ETH staked by solo stakers. If 50% of the 4 million switches to LSD applications and of that, 35% goes to Rocket Pool, that’s 721,155 additional ETH that could be staked through Rocket Pool. There’s currently nearly 430,000 ETH staked via Rocket Pool. Converting solo stakers to Rocket Pool stakers could drive a 168% growth in the amount of ETH staked via Rocket Pool as the table below illustrates.
Sounds compelling, but is this priced in?
Rocket Pool will benefit from significant tailwinds. But are these benefits already priced in by the market? How can we value the RPL token?
There are two ways of valuing the RPL token: buying demand and steady state value.
Buying demand
The RPL token is a utility token. There are no fees charged by Rocket Pool. No fee generation accrues to RPL token holders. Instead, RPL is needed to use Rocket Pool. Validators need to buy RPL as collateral.
RPL buying demand is inferred by estimating the total amount of ETH staked, Rocket Pool’s market share and the average Rocket Pool collateralization ratio. The table below outlines those assumptions. Assuming ETH staked increases to 30% and Rocket Pool’s market share rises to 8%, there is a 6.7x growth in the amount of ETH staked through Rocket Pool. At that stage, the Atlas upgrade will have gone live, which reduces the amount of ETH node operators need to contribute to a pool from 16 to 8. Assuming the collateralization level reduces to 55% from 73% today, node operators need to buy an incremental 240,891 ETH worth of RPL tokens as collateral. That’s worth $482 million at today’s ETH price. $482 million of RPL buying demand represents 49% of RPL market cap and 32% of its fully diluted market cap.
That’s a lot of buying demand!
This methodology allows us to quantify the potential buying demand for RPL. It’s likely large, but the approach falls short of inferring what the value of RPL is.
Steady state value
The steady state valuation approach values RPL once growth have leveled off. It makes the same market share assumptions as the buying demand approach and implies the total amount of RPL collateral denominated in ETH for a given set of assumptions. That value is then grossed up to imply the Rocket Pool market cap denominated in ETH. The Rocket Pool ETH denominated market cap is divided by the RPL supply to infer the price per RPL token in ETH. The price in ETH per RPL is then converted to USD.
The table below outlines these calculations. Assuming 30% of ETH is staked and 8% share for Rocket Pool with a 55% collateralization ratio, at a current ETH price of $2,000 the RPL token price should be $75.
The inferred RPL price is highly sensitive to staking ratio, market share and collateralization ratio. Notice that as the amount of ETH staked on Rocket Pool increases, but the collateralization ratio declines, the implied RPL price declines. That’s because as the collateralization ratio declines, node operators are not buying as many RPL tokens.
We may see a run up in RPL price but eventually, if the collateralization ratio settles at a much lower rate, the RPL token price doesn’t have much upside.
Rocket Pool value levers
There are two things that drive value for Rocket Pool and the RPL token. If an investment in Rocket Pool does not work out, it’s most likely because one of these two things didn’t pan out. They are:
i) rETH liquidity and usefulness
The derivative part of LSD is critically important. The reason Lido has a dominant 32% share is because its derivative, stETH, is liquid and useful. stETH’s TVL is $13 billion. stETH’s average daily traded volume is $17 million. Owners of stETH can do virtually the same thing with it as they could with ETH.
That’s not the case with rETH, Rocket Pool’s ETH derivative. rETH’s TVL is $1 billion. Its average daily traded volume is $4 million. rETH is not as useful and less liquid than stETH. There’s just not as much to do with rETH and it’s harder to trade.
Depositors who contribute their ETH to an LSD application prefer getting a useful and liquid derivative in return. All else being equal, depositors rather enjoy the greater functionality and liquidity of stETH over rETH.
If prospective ETH staking depositors rather use Lido, then the node operators will flock to Lido. It will be a winner take all market.
Why bother with a second tier ETH derivative that has limited functionality and liquidity?
Coinbase’s recent success with its LSD cbETH suggests it’s not a winner take all market. cbETH was launched in August 2022. It has grown TVL to $2.5 billion and has an average daily traded volume of $7 million.
There are two issues at hand. Liquidity and usefulness. Usefulness is determined by how many DeFi applications the LSD can be used on. Integrating LSDs into DeFi applications is not a challenge. Liquidity is the challenge. If there’s no market for the LSD, it can’t be used for much.
The ability of Rocket Pool to drive liquidity for rETH will determine the usefulness of rETH. If rETH is not useful, then ETH staking depositors won’t use Rocket Pool, which means there are fewer node operators and less RPL buying demand.
ii) Rocket Pool collateralization
The collateralization ratio is the value of the RPL tokens node operators post as collateral divided by the ETH node operators need to contribute to the pool. Node operators need to collateralize between 10-150%. For example, a node operator contributes 16 ETH, they need to post 1.6 to 24 ETH worth of RPL tokens. That’s what creates the RPL buying demand. The average collateralization ratio today is 73%. Currently node operators collateralize 12 ETH (16 ETH x 73%) worth of RPL tokens.
The higher the collateralization ratio, the more RPL token demand buying there is. Buying demand evaporates and turns into sell pressure when the collateralization ratio declines. It’s highly reflexive.
Node operators have the best unit economics at 10% collateralization ratio. They need less capital upfront. It only makes sense to invest in more RPL and increase the collateralization ratio if the node operator thinks the RPL token price is going to increase. The more RPL is collateralized the more new RPL tokens the node operator is issued. The higher the collateralization ratio, the more the node operators' economics are driven by an increase in RPL price as opposed to Ethereum staking rewards.
The table below illustrates the Steady State RPL valuation holding all variables constant except for collateralization ratio. I use the same 30% staked ETH and 8% Rocket Pool market share as previously, but show three collateralization scenarios. Current assumes the current collateralization ratio of 73%. Min assumes 10% and max assumes 150%. Note the wildly different implied RPL price based on different collateralization assumptions.
So what?
I own a small position in RPL. The thesis is compelling. The catalyst is playing out. The usefulness of rETH relative to other LSDs is debatable. So far it does not seem to be hampering the growth in ETH staked via Rocket Pool (see table below).
RPL’s current valuation is less compelling. The token price is already pricing in a significant increase in ETH staked via Rocket Pool. I don’t think the market understands the reflexivity of the collateralization ratio. To my surprise, the collateralization ratio keeps increasing. I closely monitor growth in ETH staked via Rocket Pool and the collateralization ratio to assess when position sizing.
My research on LSDs is ongoing. Please let me know if you agree or disagree with the points I have made.
Stay curious.
Don’t forget to hit the “♡ Like” button!
♡ are a big deal. They serve as a proxy for new visitors and feed into Substack’s algorithm that distributes my articles to all Substack readers.
While the growth case is clearly compelling, LSD appears to be suffering from 2 fronts - 1. dozens of new entrants; and 2. commodity offering leads to margin pressure. While there's market share to be gained, margins will start to race to the bottom as competition increase. I suspect momentum might be short lived. Thoughts?
great article ser. Learnt a lot !