Discover insights on Ethereum staking dynamics, liquid staking, and institutional strategies from executives at Amberdata in an interview as part of Moody's Bits, Bytes & Basis Points series. Learn about the impact of Ethereum upgrades, liquid staking tokens, macroeconomic factors, and DeFi lending protocols in the crypto staking space.
Staking is a pledge of capital to a blockchain by owners of a specific cryptocurrency. In a “proof-of-stake” consensus mechanism, stakers enhance the security of the associated blockchain network by having a vested financial interest in it and participating in validating transactions, and they can also get a return on their investment. Following the Ethereum blockchain's transition to a proof-of-stake consensus mechanism powered by staking of ETH, its native token, the dollar value of ETH staking has grown to more than one third of the total value locked in the decentralized finance (DeFi) ecosystem. There are now many investment possibilities for institutional and retail investors that interact with ETH and their staking derivatives. In our view, growth of the ETH staking market also poses multiple risks related to the complex technology stack that powers staking applications. Market and liquidity fluctuations could also have widespread repercussions in digital finance markets given the high interconnectedness between staking and the broader DeFi ecosystem.
We asked Pat Doyle and Greg Magadini of Amberdata about how ETH staking is being implemented in the industry. Their responses are set out in question-and-answer format below and express their own views, not those of Moody's. Below is a summary of key points from Amberdata's responses to Moody's questions.
Ethereum's Merge and Shapella upgrades are its most significant. The number of validators has continued to increase as a result of these two important changes.
Liquid staking has proved to be extremely capital efficient. The adoption of liquid staking tokens across the entire crypto market has reached $44 billion, and the introduction of restaking could further enhance the growth of liquid staking.
Liquid staking has given rise to “delta-neutral” investment strategies. As these strategies become more common, their returns will converge toward those seen in traditional finance.
Liquid staking has created some centralization around pools of ETH controlled by single protocols. Although research is underway to address this issue, participants to the Ethereum ecosystem need to carefully assess the implications of more centralization.
How has the staking yield on ETH evolved following the Merge and Shapella upgrades, and what factors contribute to its attractiveness for investors?
Pat Doyle and Greg Magadini (Amberdata):
Ethereum has undergone many updates that have changed the issuance of Ether, its native currency. Before we dive deep into the Ethereum staking environment, let’s take a step back and review how Ethereum got to a proof-of-stake consensus mechanism.
We have two main time periods to consider. Pre-Beacon-Chain2 and Post-Merge. Pre-Beacon-Chain refers to the period where Ethereum still operated under a proof-of-work consensus model. Post-Merge is the current period when the Ethereum network switched to a proof-of-stake consensus mechanism.
Below is a table of major Ethereum events that have impacted the ETH supply over time
The Beacon Chain upgrade allowed ETH holders to begin staking. The Merge, when Ethereum’s original proof-of-work blockchain merged with the new proof-of-stake Beacon Chain, was where we saw the supply dynamics shift away from a static block reward (a fixed number of ETH per new block created) to a dynamic reward, based on the amount of ETH staked and other factors. Lastly, the Shapella update allowed stakers to remove their ETH from the Beacon Chain. This is fundamentally important because the number of validators on the Beacon Chain has a direct effect on ETH staking yield.
The Merge is considered the most significant update to the network and was part of the original Ethereum vision. This update from proof-of-work to proof-of-stake changed the way transactions were validated on the network from a system of rewarding participants for “mining,” or running computationally expensive processes, to choosing validators randomly among participants who had staked the required 32 ETH, minimizing the network’s electrical output. Proof of work mining is also a capital intensive operation, but proof-of- stake offers validators an easier access point into securing the network.
The ability to begin staking ETH on the Beacon Chain began in late 2020, but at that time did not include the ability to withdraw that staked ETH. This meant that validators were locking their ETH up in the Beacon Chain until the Shapella upgrade went live.
Because stakers could not withdraw their ETH, the annual percentage yield (APY) of ETH staking before the Shapella upgrade remained a function of the amount of validators entering the Beacon Chain. See Exhibit 2 below: as more validators enter you can see the APY decrease, leading to a consistently decreasing APY. In early 2021, staking yields were in the low teens and trending down as more validators came online, ending at around 3.5% at the time of the Merge. For a more technical overview on Ethereum issuance post Merge, please see this write-up.
The Shapella update allowed validators to withdraw their ETH from the Beacon Chain. This is important because as mentioned earlier, the ETH yield is a function of the number of validators. The greater the number of validators, the lower the yield, and vice versa.
Now that stakers have the choice to remove their ETH from the Beacon Chain and effectively stop staking, have we seen a change in their staking behaviors?
In fact, the number of validators continues to increase post Shapella. So as a result, we continue to see the ETH staking APY decrease. The current ETH staking yield is around 3% APY.
The ETH staking environment is still relatively young but the market will likely ultimately find a point of equilibrium between active validators and ETH yield.
Can you elaborate on the impact of the introduction of Liquid Staking on the dynamics of staking yield for ETH, and how does it add complexity to the overall ecosystem?
Pat Doyle and Greg Magadini (Amberdata):
Staking Ethereum comes with a few requirements for stakers. First, you need to stake exactly 32 ETH to run a validator, and also must have the ability to run and maintain both an execution layer client and a Beacon Chain client. These two requirements have either priced out holders of smaller ETH amounts or have made staking difficult for ETH holders with less technical knowledge.
These two obstacles have led to the creation of a new market of Liquid Staking Tokens (LSTs). Liquid staking is the process of depositing your ETH to a liquid staking protocol, the protocol issues the user a LST and the protocol then stakes on the users behalf and for this service they take a portion of the rewards.
The objective of LSTs are to enable users to gain rewards from deposits under 32 ETH and to allow for deposits that are not strictly multiples of 32 ETH. They allow the liquid staking token to be used for various applications and protocols, for example serving as collateral in lending or other DeFi initiatives which creates a more capital efficient staking environment. Additionally, they offer an alternative to exchange staking, self-staking, and various semi-custodial and decentralized protocols.
Below is a diagram of how liquid staking protocols operate at a high level.
For offering these services the protocols will take a portion of the fees they accrue for staking on behalf of the user. Staking yields for LSTs are therefore slightly lower than yields for staking ETH directly. However, because capital efficiency is much higher and capital requirements are less onerous for staking LSTs, we have seen adoption of LSTs across the entire crypto market. Currently, the market capitalization for liquid staking derivatives is $44 billion.
Let’s now do a deeper dive into LSTs, their dynamics and their yields. For this analysis we will cover the larger LSTs on Ethereum, which are Lido Protocols-stETH, Rocket Pool-rETH and Coinbase-cbETH.
Each of these protocols provides a similar service to somewhat different audiences.
» Lido - stETH is the pioneer in the industry for liquid staking. This LST offers the lowest fees and no minimum staking requirement, but has the highest centralization among liquid staking providers.
» Rocket Pool-rETH has slightly higher fees and offers a more decentralized staking network.
» Coinbase-cbETH has the highest fees, but also the easiest onboarding because retail traders do not have to move funds off of the exchange and can passively enable staking while holding ETH on Coinbase.
Below, look at LSTs by fees and market capitalization.
Exhibit 5
Main liquid staking tokens
Looking at on-chain data we can begin to analyze the average 30-day average yield of each LST.
Why is there a difference in yield among different LSTs? First, let's look at the fees charged by the leading LSTs. Above, you can see that Lido not only is the leader in terms of market capitalization but also charges the lowest fees. These fees have a direct impact on the users' staking yield.
Looking at market capitalization of these assets we can see that Lido stETH has been live since early 2021. This has given them an advantage as a first mover in the market. However, we have not seen other LSTs come in and take market share from stETH. Being a first mover and having the lowest fees has helped Lido remain the leader by market capitalization in this space.
How do shifts in macro-level interest rates impact institutional crypto staking strategies, particularly in terms of asset allocation and risk management for large-scale staking operations?
Pat Doyle and Greg Magadini (Amberdata):
Yields on short-term government bonds, often used as a proxy for “risk-free” rates, are the benchmark opportunity cost for investment decisions worldwide. A higher risk-free rate can reduce the attractiveness of investments in cryptocurrency, because cryptocurrency
is much more volatile than government bonds. Although US short-term interest rates are the highest seen since the initial launch of the BTC white paper, rates are different around the world. Rates in Japan continue to be around 0%, while elsewhere in Europe the increases have not been as aggressive as in the US. As a result, globally, the appetite for cryptocurrency investments has mixed.
Another more important point however is in the implicit rate environment within the crypto sector. Perpetual funding and the future’s “basis,” which are implicit interest rates reflected by leveraged crypto derivative instruments, show rates that far exceed global risk-free rates. For example, as of this writing, the Bitcoin futures basis for 90-days current has a 25% annualized cost associated with it, compared to 5.25% annualized risk-free. This shows a large dislocation between the cost of capital for leveraged Bitcoin investing, suggesting that the US risk-free rate has almost no impact on the Bitcoin investment decision process.
Another source of yield within the crypto space is rewards from mining, which, as discussed earlier, allows participants to validate transactions. Traditionally Bitcoin mining has required the tradeoff between hardware investment costs and the subsequent stream of Bitcoin mined as revenue. However, now that Ethereum has created a proof-of-stake consensus mechanism, the investment required has shifted from hardware to native tokens, which then can be staked for yield.
Holding large investments in ETH, which can be staked for yield, also allows investors to sell covered calls against their holdings, further enhancing passive yields. This is analogous to investing in high dividend stocks and selling covered calls, a strategy often used in traditional markets.
How does the maturation of the crypto market influence the decisions of institutional crypto stakers engaging in basis trading? Specifically, how do shifts in implied yield from futures contracts impact their staking strategies?
Pat Doyle and Greg Magadini (Amberdata):
Holding large investments in stakeable assets, such as ETH, enables investors to write covered calls to enhance yield, but this strategy comes with spot price investment risk. Combining the futures basis and the staking yield together in a portfolio actually enables investors to isolate the total yield from the investment without being exposed to ETH price risk. Investors can do this by first buying ETH, staking the ETH for, say, 5% yield, then selling a corresponding future against their ETH holdings, which yield around 25% as of this writing, for a combined total yield of 30%. Any drop in the price of ETH will be offset by the short future position, although any gain in spot price will also be offset. This “delta neutral” 7 position yields significantly higher returns in crypto than similar structures found in traditional markets.
As the crypto market matures and these staking strategies are used more often, we would expect the total return to converge from the currently high 30% annualized returns down to more reasonable levels approximating those of traditional finance, and eventually down to risk-free rates.
Considering the evolving landscape of DeFi lending protocols, how are institutional investors using their crypto staking assets outside of just generating staking yields?
Pat Doyle and Greg Magadini (Amberdata):
In general, LSTs are being used in a few different areas of the DeFi ecosystem:
EigenLayer9 is a protocol that has introduced “restaking.” Restaking is the idea that you can use the security of Ethereum to secure other networks as well. This is done by essentially bonding your LST to the EigenLayer contract to secure other protocols. In exchange for restaking it is anticipated that restakers will receive protocol fees and rewards for providing this security.
Currently, there is 3,500,000 ETH/LST staked in the Eigen layer, which is roughly $12,000,000,000. There is no direct yield here for staking at the moment.
Below you can see the distribution of the LSTs being restaked within EigenLayer.
Read the full report here as this is just a subset of the report!