July 4, 2023
 min read

Shanghai and Capella: Understanding the Ethereum upgrades and how they will affect the network

Thomas V. Daugaard

In September 2022, Ethereum achieved a significant milestone when it switched the consensus mechanism from Proof-of-Work (PoW) to Proof-of-Stake (PoS). This change resulted in a drastic reduction in energy consumption by about 99.95% and paved the way for scalability updates that were previously impossible under PoW.

The next step on Ethereum’s roadmap is the Shanghai and Capella upgrades, which signifies the final phase of Ethereum’s PoS vision. In brief, the upgrades will enable the withdrawal of staked ETH and are expected to take place on the 12th of April 2023.

Here is a comprehensive breakdown of what we can expect from the upgrades, the withdrawal period, and what it means for investors.

The results of a successful merge

During the Merge, the Ethereum mainnet was integrated with the Beacon chain, which is the PoS chain launched in 2020 to test the novel consensus mechanism.

With the merge, the computational operations that consume significant energy to validate transactions in the traditional PoW system are now obviated. Instead, network participants pledge 32 ETH into a smart contract to become validators, who secure the network and verify transactions, in exchange for newly minted ETH as rewards.

As a result of the switch from PoW to PoS, the costs that the protocol paid miners for validating transactions and providing security, decreased significantly. Before transitioning to PoS, miners were issued approximately 13,000 ETH/day, whereas stakers today are issued approximately 1,800 ETH/day, as shown by token terminals’ overview of token incentives paid by the network, which can be seen below.

Source: Token terminal

Following the decrease in the issuance of tokens, the inflation rate consequently dropped. With ultrasound.money’s estimated annual issuance of 657,000 ETH and a total ETH supply of 120,455,000, the protocol yields an annual inflation rate that is 0.55% down from 4.61% pre-merge. In addition, however, due to the burning mechanism introduced by EIP-1559, the supply growth of ETH has become negative — essentially making Ethereum net-deflationary.

In several ways, this is beneficial for token holders staking their ETH as it implies that the real yield supersedes the nominal yield — a first among all PoS blockchain protocols.

To date, more than 550,000 validators have staked over 17.5 million ETH, which equals almost $30 billion. However, they have not been able to withdraw tokens, and will not be able to until the completion of the Shanghai and Capella upgrades.

What is the Shanghai and Capella upgrade?

Over the last two years, Ethereum network validators have voluntarily locked their ETH on the Beacon chain without a timeline for when a withdrawal function will be added. The upcoming Shanghai-Capella upgrade targeted on April 12th will put an end to this era by allowing validators to access their staked ETH, alongside the rewards they have earned for their validation activities.

Capella will upgrade the Beacon chain, the consensus layer, while Shanghai targets the execution layer formerly tied to PoW. The indications are positive for the progress of the implementation. A state copy of the mainnet has been successfully created through a shadow fork, and staking withdrawals are already processed on the Zhejiang, Sepolia and Goerli testnets, marking the end of the testing phase.

The Shanghai fork will implement EIP-4895. This improvement proposal introduces a system-level operation to support validator withdrawals that are “pushed” from the beacon chain to the Ethereum Virtual Machine. This means that a new type of transaction will be implemented that pushes withdrawals from the consensus layer to the execution layer.

The Capella upgrade is the upgrade necessary for withdrawals on the consensus layer. The process required to carry out a full withdrawal occurs on the consensus layer, while the execution layer is responsible for processing withdrawal requests from the consensus layer and transferring the ETH to the correct address.

Capella will facilitate two main types of withdrawals: full and partial. A full withdrawal involves a validator’s request to exit the active validator set and is irreversible. After a certain period, the validator’s staked ETH and earned rewards become withdrawable and are picked up by the withdrawal sweep mechanism, which sends them to the validator’s withdrawal address.

Partial withdrawals will be automatically executed, with any amount over 32 ETH being sent directly to the validator’s designated withdrawal address. As validators earn consensus rewards, their balance may exceed 32 ETH, but any amount above this threshold will not earn the validator additional rewards. For this reason, the protocol will automatically withdraw amounts over 32 ETH.

Essential elements of the withdrawal process

It is important to note that there are limits for both partial and full withdrawals. In order to maintain a balance between clearing withdrawal queues and avoiding blockages caused by withdrawals, the development community has determined that the number of withdrawals allowed per block will be limited.

If a validator intends to completely exit the Ethereum network and withdraw both their stake and the consensus layer rewards, they must submit a request to exit the active validator set. It is mandatory to wait for a minimum of four epochs (approx. 25 minutes) before the request is processed. As of now, a maximum of seven validators are allowed to exit per epoch, which means that validators must wait their turn in case of a queue. During this time, validators are still obligated to perform their duties.

After exiting the active set, the validator must wait for an additional 256 epochs (approx. 27 hours) if the validator has not been slashed, and 8,192 epochs (approx. 36 days) if slashed. This large discrepancy is intended to disincentivize bad actors. Once this waiting period is over, the validator’s stake and consensus rewards can be withdrawn.

Given that a maximum of seven validators are allowed to exit per epoch, the actual withdrawal period will depend on the validator churn. According to Blockworks, if 25% of the entire validator set were to try and exit in one day, it would take at least 80 days to complete. And to expect the same withdrawal time as e.g. Cosmos’ chains (21 days) would require between 6.3% and 7.2% of the validator set to be in the exit queue at one time.

Source: Blockworks

What implications do the upgrades have on the network?

The Merge significantly contributed to Ethereum’s value proposition and provided a boost of confidence for the implementation of its ambitious roadmap. As 2023 unfolds, we anticipate a flurry of upgrades, starting with the Shanghai and Capella upgrade that will allow staking withdrawals. As we move into Q2 2023, it will be interesting to see how the market will react to the influx of unlocked ETH. In the following analysis, we explore three elements that may affect the potential selling scenarios.

ETH is already liquid

The first element that may affect the price development of Ethereum after the Shanghai and Capella upgrades is the presence of liquid staking derivatives, such as Lido’s stETH.

These derivatives provide a liquid option for users and investors who wish to exit the validator set. As of now, the majority of staked ETH is held by a few liquid staking providers, such as Lido, alongside centralized exchanges like Coinbase, Kraken, and Binance. In addition, approximately 25% of staked ETH is held by unidentified stakers, primarily composed of solo stakers or smaller groups running private staking pools. The distribution can be seen on Nansen’s dashboard below:

Source: Nansen — ETH2 Deposit Contract

This implies that more than 50% of all ETH deposits are already liquid through decentralized or centralized providers, while 25% is staked with solo stakers, who are normally long-term holders with large investments in node setup. As such, around 75% of the overall stake is either liquid or solo staked.

Therefore, it is reasonable to assume that the influx of unlocked ETH from staking withdrawals will have an insignificant impact on the price development of Ethereum due to the current distribution of staked ETH and the availability of liquid staking options.

Most validators are underwater

The second element that may affect the price development of Ethereum after the Shanghai and Capella upgrades is the cost basis for most Ethereum validators.

As of now, the ETH/USD price at the time most validators staked on the Beacon Chain is above the current exchange rate. This means that about 80% of validators have little financial incentive to sell their ETH, as their cost basis is higher than the current exchange rate. Only around 27% of validators have a lower cost basis and thus have potential unrealized profits — as can be seen in the illustration from Dune below. However, these validators are also the ones with the highest conviction, as they were bold enough to enter the validator set early and are therefore considered unlikely to sell their ETH.

Source: Dune — Ethereum Staking

This highlights the likelihood of only moderate to insignificant sell pressure from validators exiting the validator set. On the other hand, some selling pressure may be induced by validators being underwater for tax loss harvesting purposes, especially as the deadline for tax returns in the US expires in April 2023. Nonetheless, given the current distribution of validators and their cost basis, it is unlikely that the influx of unlocked ETH from staking withdrawals will have a significant impact on price development.

The staking rate of Ethereum remains the lowest among L1s

The third element that may affect the price development of Ethereum after the Shanghai and Capella upgrade is the staking rate of Ethereum.

Despite the steady growth in deposit activity and revitalized sentiment toward the upcoming Shanghai and Capella upgrades, Ethereum’s staking ratio remains relatively low. Although it’s two years since the Beacon chain launch, only 16% of the ETH supply is staked, compared to other smart contract platforms that show significantly higher ratios e.g., Solana at 71%, Cardano at 69%, and Cosmos at 61%.

This is likely due to the fact that staking ETH required a risk profile willing to lock a risk-on asset for an indefinite period, which deterred some investors and institutions. As Shanghai and Capella de-risk staking and transform ETH into a liquid asset that generates yield, it is reasonable to expect inflows from investors and institutions that were previously hesitant to stake their ETH.

This influx of previously side-lined institutions will lower the free-floating supply and could even create demand, which may have a positive effect on the price development of Ethereum.

Concluding remarks

Overall, while there may be some selling pressure induced by tax loss harvesting or early validators realizing profits, the presence of liquid staking derivatives and low financial incentives to sell suggest that any selling pressure will be moderate to insignificant. Meanwhile, the expected inflows from investors and institutions as staking is de-risked could drive demand and further boost the price development of Ethereum.

At Northstake, we simplify the way institutional investors own a stake in future web3 economies. To find out more and start staking ETH with Northstake, contact us at: info@northstake.dk or visit www.northstake.dk

Thanks for reading and remember to stay up to date by following our Medium account or by joining our community below.