In just two years, Ethereum-based liquid staking has become the largest asset class on-chain, eclipsing all decentralized exchange TVL in April 2023. Liquid Staking Tokens (LST’s) is the hottest narrative in crypto right now with hundreds of different teams working on different versions and integrations of them into DeFi.
In this guide, we're going to start with the basics of staking and why LST's were created. Then were going to look at the different options for liquid staking. Last, we’re going to show you why all LST markets are just lending markets and share recently released information about the upcoming design of frxETH v2, Frax’s near perfect model for creating ETH-validator lending markets.
What is Ethereum Staking?
Staking is a mechanism used in Proof of Stake (PoS) blockchains where participants lock up their tokens to secure the network and validate transactions. For the Ethereum network, it means depositing 32 ETH to activate a validator. Validators are responsible for processing transactions, storing data and adding new blocks to the blockchain.
Validators keep the Ethereum network secure.
The point of staking is to create both monetary incentive and disincentives for validators to process blocks properly. Good node operators earn ETH, bad operators lose ETH. It's very simple.
If a validator, or a group of validators ever acts in a malicious way, their stake is "slashed" and they lose part of their ETH.
As a reward for properly validating transactions new ETH is distributed to node operators.
What is “The Merge” and when did it happen?
In December 2020, the "beacon chain” was launched in parallel to the main Ethereum blockchain. At the time the beacon chain only allowed users to stake ETH, no other functions were allowed.
On September 15, 2022, the Ethereum network successfully executed “The Merge,” transitioning from energy intensive Proof-of-Work to environmentally friendly Proof-of-Stake. The Merge change the way that transactions are validated. Previously, Ethereum’s consensus mechanism was similar to Bitcoin; millions of computers competed to solve a mathematical puzzle, called a hash, and the winner was rewarded with new token issuance. With the merge complete, transactions are now confirmed by validators staking ETH, processing transactions, and earning a reward for their services.
What are the ways I can stake my ETH?
There are four main ways that you can stake your ETH:
- Centralized exchanges
- Solo staking
- Pooled staking aka LSTs
Centralized exchanges (Binance, Coinbase, Kraken)
Up until recently, one of the most popular ways to stake your ETH was on a centralized exchange. Well known CEX's like Binance, Coinbase, and Kraken all had 1-click easy to use staking solutions. Simply buy some ETH through their exchange and then immediately stake using their custodial services.
These centralized offerings were wildly popular, Coinbase’s Wrapped Staked ETH (cbETH) topped out with more than 1.25m ETH staked, Kraken at 1.3m and Binance at 1.1m.
The problem, unfortunately, is that while these centralized venues provide an easy way to stake your ETH, the trade-off here is that they have high trust assumptions and, more importantly, are a huge target to regulators.
The SEC is on the warpath to shut down all staking in the United States. In February 2023, the US regulatory agency settled with Kraken, who shuttered its program for US customers and paid a $30 million dollar fine. And in May, they filed suit against Coinbase and Binance, claiming their staking programs are unregistered securities offerings.
With no clear regulatory laws in the United States, staking through a centralized exchange should be considered one of the riskier methods covered in this guide.
With solo staking, you host the equipment, maintain the software and also stake a full 32 ETH. Solo staking is by far the most decentralized and secure way to support Ethereum network security.
But solo staking does have its risks. If you cannot effectively maintain good uptime, the software or hardware, you could be putting your principal at risk and get slashed.
These are third party companies that handle all of the hardware and software maintenance for you, so that you only need to deposit your assets. These services are typically run for those who don’t want to solo stake, but don’t want to swap their assets into a liquid staking token.
Pooled staking aka Liquid Staking
The simple description of liquid staking is that you swap your ETH for another ERC-20 token, the LST, which then captures the staking reward yield generated by a set of node operators.
In short, an LST is a loan made by ETH holders to node operators.
Whenever you swap your ETH for an LST, you are effectively loaning your assets to the validator node operators. You trust them with your loaned assets to provide sufficient security guarantees, exceptional operational uptime and interest paid in the form of rewards yield. The loan is closed when you redeem the LSTs. If validators are slashed, the loan collateral is reduced globally for all token holders, so it’s imperative that LSTs performance is world-class.
There are two primary ways that the reward yield and slashing conditions can be transmitted to token holders, the price of the token can be adjusted or the supply can be rebased. DeFi prefers number go up, so even if a token rebases, it’s typically wrapped (wstETH) for wider usage.
Outside of these differences in liquid staking token mechanics, the trust assumptions, bond requirements and security of the validator set comprise all major variations between protocols. Let’s look at Lido’s stETH, RocketPool’s rETH and Frax’s frxETH to see how each operate.
stETH (Lido staked Ethereum)
Lido is a liquid staking protocol launched in 2020 that supports Ethereum staking and other Proof-of-Stake networks. Lido is currently the largest staking protocol by TVL, with over 6m ETH staked.
When you use Lido, you are swapping ETH for stETH, Lido’s liquid staking token. You send your assets to their minting contract, and then a set of node operators borrow your ETH and stake it on your behalf.
All of Lido’s validators are public, centralized, have undergone KYC/AML checks, signed legal contracts with the DAO and have an off-chain relationship with Lido.
Use the loan analogy as a lens, Lido node operators are able to borrow ETH with zero collateral requirements, with the strict assumption that all ETH rewards earned from staking, minus fee, are returned to lenders (stETH holder) as interest through rebasing.
rETH (Rocketpool Ethereum)
RocketPool is a decentralized Ethereum staking pool founded in 2016. RocketPool pairs people with less than 32 ETH with node operators who stake on their behalf.
When using RocketPool, you can either swap ETH for rETH, their native LST, or you can borrow ETH from the protocol and be a node operator.
When you swap ETH for rETH, a 0.05% fee is deducted by Rocket Pool. The conversion rate is determined based on your proportional entitlement to the ETH in the Rocket Pool protocol.
Node operators are decentralized and anonymous. In order to create security guarantees, node operators must post either 8 or 16 ETH as collateral. Additionally they must also purchase Rocketpool’s token RPL and then bond it as collateral. In case of slashing, the RPL is sold to backstop the loan collateral.
Rocket Pool operates with a fixed 15% commission rate that provides node operators with a share of the rewards earned on the ETH assigned to them by the protocol. This setup allows node operators to earn rewards on their own ETH deposit as well as a commission from the loaned ETH.
Rocket Pool node operators have a higher LTV than Lido competitors and also charge higher fee to rETH stakes. Node operators must post 8 or 16 ETH and a minimum RPL value of 10% compared to collateral. Due to the higher LTV, RocketPool yields almost always are lower than Lido’s.
frxETH & sfrxETH (Frax Ethereum)
The core mission of the Frax Protocol issues cutting-edge, decentralized stablecoins and incorporates supporting subprotocols to bolster their functionality. Currently they issue 3 stablecoins, FRAX, FPI, and frxETH. They are pegged to the USD, the US CPI, and ETH, respectively. We’re going to focus on the latter in this article.
Taking our description from our Frax 101 article on frxETH:
frxETH and sfrxETH are the two tokens comprising Frax’s liquid staking system.
Frax Ether (frxETH) is an ETH-pegged stablecoin. It receives no rewards from staking and it always should remain highly correlated with ETH. As a comparison, it is most similar to Wrapped Ether (WETH).
sfrxETH is frxETH that is staked in a ERC-4626 vault. sfrxETH receives all of the staking rewards earned by validators. sfrxETH is non-rebasing and its price goes up slowly over time as rewards are earned.
In Frax’s system, swapping ETH for frxETH, you loan your assets to the Frax protocol, which are then staked in validator nodes that the Core Dev Team currently manages. As the nodes are currently centralized, no collateral or tokens are required to bond
Loan interest is only paid out to sfrxETH holders. frxETH receives no native incentives. Instead, frxETH can earn rewards from other DeFi protocols, most notably Curve where Frax uses its CRV/CVX power to vote for its pools.
When a frxETH holder wants to close the loan and claim their ETH, they simply sell the frxETH into the Curve LP.
A Loan By Any Other Name
In all three of these cases, the general concept behind every model is that ETH is lent out to node operators who stake and earn rewards. Interest is passed back through to LST holders in the form of staking rewards. The only difference between all three protocol is the LTV of the node operator, the trust assumptions of the protocol, and the security guarantees provided by the node operators.
Given all this, the best mental model for liquid staking as a protocols is that they create an ETH borrowing market for node validators.
The best LST protocol is the one that provides:
- Best rates for borrowing ETH.
- Highest decentralization, strongest trust assumptions and security guarantees from node operators.
- Deepest swap facility for trading
- Optimal token construction for integration into DeFi
Based on the above conditions, Frax designed frxETH v2 to excel above every other LST protocol. Let’s dig into how it works.
How frxETH V2 Sets Itself Apart
frxETH v2 creates a decentralized, agnostic, peer-to-pool ETH lending market for node operators.
The best way to think of it is in the context of am isolated lending pool, similar to Fraxlend or Aave, where node operators can borrow ETH at dynamic rates to fund their validators. Node operators do not charge any fees or commissions.; the only fee is what the lending market charges.
If the ETH borrowing rate is low (low ETH utilization in the lending pool), node operators are economically incentivized to borrow ETH at sub-market rates relative to the staking rate, arbing a higher spread. They pay interest to sfrxETH while keeping all MEV profits, tips etc… If the interest rates jump up to the point of unprofitability (high ETH utilization) the node operators can eject from the beacon chain, repay their debt, and then wait for an interest rate that is more to their liking.
Dynamic interest rates reward the best performing validators who can earn the highest amount of yield relative to the borrowing rate. The best performing validators will draw the most economically optimal amount of ETH to maximize their profits. Subpar validators are disincentivized, as they will be subjected to higher borrowing costs and potential operating losses.
LIQUIDATION!! EJECT! EJECT!!
Node operators that borrow ETH as collateral for their validators enter into an explicit loan with the protocol.
When taking the loan, the node operator must register their validator public key and set the withdrawal addresses to the protocol lending market. This ensures withdrawn funds can only go to a Frax protocol controlled address once ejected.
As with any loan, node operators are required to maintain a specific Loan-to-Value ratio. If the validator engages in malicious behavior and is slashes, its LTV will be reduced and eventually enter into liquidation. At this point, the Frax protocol forces a full withdrawal to eject their collateral back to the lending pool. Node operators may always top up their collateral amount in the lending market to remain healthy.
As mentioned in our previous article Frax 101: frxETH and sfrxETH:
In its current model the Frax Core Team completely controls all validator nodes using a 3/5 multi-sig. While this has been necessary for initial growth and launch, now at a more mature state this control is dangerous and threatens the health of the entire Frax protocol.
Once frxETH v2 is live, it will allow any validator to borrow ETH from the lending market. In turn, this will fully decentralize the frxETH system and enable it to scale and grow in a healthy manner without placing excess risk on the Frax multi-signers.
To borrow a validator, a node operator must provide a minimum amount of collateral (for example, 8 ETH) to gain access to the lending market. As mentioned before, the ETH is constrained to only be staked in validator nodes that set their withdrawal address to the Frax lending market. Borrows can’t just post 8ETH and then run off with 24 ETH, the 32 ETH can only be used in the protocol issued validator.
The beauty of this system is that none of the node operators have to be trusted and can remain anon. Trust assumptions about the security of the ETH are instead held by an oracle system that tracks the health of all loans.
Currently, it is not possible to read Ethereum’s Beacon Chain state using smart contracts. Because of this, the use of an oracle is necessary to update the performance of the validators. Frax’s Beacon Oracle uses ZK technology to prove that the computations are actually valid. If a validator is ever found to have an unhealthy loan, the oracle will update and eject the collateral back to the lending pool.
So there is a trade off. Instead of trusting node operators not to run off with your capital or get slashed, you instead have to trust that the oracle system is able to maintain loan balance information across the network.
How Idle ETH Goes to Work in the Curve AMO
Any ETH that is left idle in the frxETH v2 lending market is sent directly to the Curve AMO contract that balances the protocol-controlled liquidity. frxETH holders can use this extremely deep liquidity pool to enter and exit back to ETH.
FrxETH V2 markets operate similar to Fraxlend markets, the interest rate is determined by the utilization rate. As more and more ETH is borrowed by node operators, the utilization rate with rise. Once it passes the “vertex utilization,” increases to rates sharply rise until they hit a max rate when full utilization occurs. The vertex utilization rate should be the natural resting point in a healthy market system.
For frxETH v2, the vertex utilization point is set so that excess ETH in the lending contract can be used by the Curve AMO.
In v1, 10% of all ETH is retained and paired with frxETH in the Curve LP. The frxETH/ETH pair has the second largest TVL on Curve for all ETH pairs and the highest liquidity profile relative to its market cap for all LSTs. The Curve AMO is unmatched at providing liquidity. Refer to our in depth guide to see how this AMO works.
What’s interesting about the market is that its self balancing between the two products. Let’s take a look at an example.
When node operators are not borrowing ETH, utilization will remain low and interest rates will drop. At the same time, the ETH/frxETH Curve LP will imbalance heavily towards ETH and the peg will drop in favor of new buyers of frxETH.
Node operators should be incentivized at this point to borrow more frxETH and sell it into the Curve LP to arbitrage the price spread.
If node operator demand is high, ETH utilization rate will be higher than the vertex utilization, an interest rates will spike. One of two outcomes will occur. First, traders will swap more ETH to frxETH and deposit into the sfrxETH contract, where yields will be higher than other LSDs. Second, node operators will eject and withdrawal their borrowed ETH and repay the loan, increasing the available ETH to borrow and lowering rates.
The whole system is designed to self balance around the natural ETH reward rate. Node operators will not take loans with interest rates that exceed their profits. Conversely, cheap ETH loans relative to rewards should always draw in new operators who want to increase their profits.
Bringing It All Together
Frax’s greatest strength is building from a first principles perspective and in the case frxETH, that perspective is “what must the rational economic incentives must be to build an LST that scales with the least possible trust assumptions”. Whether it’s the peer-to-pool lending model or Curve AMO, frxETH is leveraging primitives in the most efficient way possible in order to both bring the highest yielding product to LST holders while offering the highest upside to node operators. If everything operates as expected, node operators will naturally be lured to frxETH and the highest performing ones will be incentivized to stay. frxETH is one of several developments on the horizon for the Frax ecosystem and as the core team continues to skate where the puck is going, we are here patiently observing what they have in store for us next.