The Shanghai upgrade and what it means for investors
Mar 2, 2023 - 10 min read
The Shanghai upgrade scheduled for the first half of 2023 will enable withdrawals of both staked ETH and accrued staking rewards that remained locked in the Beacon chain. Since its genesis block, 525k validators staked more than 16.8m ETH worth $26.9b and accrued 1.05m ETH worth $1.68b in staking rewards.
The upgrade is substantially less complex than the previous Proof-of-Stake transition. As testing goes smoothly, no implementation hiccups or major delays are expected.
From an investment perspective, it is valuable to assess volatility and potential price implications around unlocked supply entering the market. Our analysis indicates minor to moderate sell pressure post-Shanghai that might offer opportunities to market participants.
The expected sell pressure is limited by withdrawal and exit queues. With 50.2% of the overall stake being liquid anyway and another 24.3% likely stemming from solo stakers that seek long-term exposure to ETH, the resulting sell pressure should be fairly moderate. Moreover, 79.5% of validators had a higher cost basis at the time of deposit and therefore no profit to realize.
We expect the first days post-Shanghai to come with elevated volatility and sell pressure, yet likely below 2% of average spot volumes, as partial withdrawals coincide with full withdrawals. Subsequently, full withdrawals will probably account for roughly the same daily ETH issuance as seen pre-Merge.
Observed in isolation, the Shanghai upgrade is a bullish catalyst for the Ethereum ecosystem as it enables more stake from sidelined entities to enter the game. We expect institutional interest to broaden as withdrawals de-risk staking, enhance liquidity and might even morph ETH’s staking rewards into crypto’s risk free rate. Its impact however might be softened by the current risk-off sentiment and uncertain macro conditions in the short term and in result come with elevated beta.
A brief technical overview
Almost 170 days ago, the Ethereum network finally transitioned from Proof-of-Work (PoW) to Proof-of-Stake (PoS). Despite the Merge being one of the most significant structural shifts of any large-scale crypto asset to date, it did not enable the full range of PoS functionalities. As such, the protocol issued inflation rewards, also known as consensus layer rewards, and the underlying stake of early validators remained locked. While the consensus layer rewards are not accessible until the upcoming upgrade, execution layer rewards like Tips and Maximal Extractable Value (MEV) are distributed ever since the Beacon chain launched in 2020. Execution layer rewards are directly proportional to the transaction activity. Technically, validators receive these as additional Tips for prioritizing, including, excluding, or reordering transactions. Consensus layer rewards on the other hand are inversely proportional to the amount of ETH staked and refer to the rewards from the issuance of new ETH. Validators receive these for participating in the security of the Ethereum blockchain either as block proposers, attesters, or members in sync committees.
Scheduled for the first half of 2023, the next Ethereum upgrade after the Merge is the highly anticipated and crucial Shanghai and parallel Capella upgrade. Capella will upgrade the Beacon chain (consensus layer) while Shanghai targets the execution layer that was formerly tied to PoW. The implementation progress appears to be promising. With a successful shadow fork, a state copy of mainnet, and Zhejiang as well as Sepolia testnets already processing staking withdrawals, the first major set of testing has commenced. Following up, Shanghai will activate on the public testnet Goerli before mainnet activation. These testnet implementations are not only important to identify glitches in the upgrade but also helpful to infrastructure providers such as staking pools or exchanges to test their tooling.
As soon as Shanghai with its most important Ethereum Improvement Proposal (EIP)-4895 (Beacon chain push withdrawals as operations) is activated on mainnet, validators aiming to exit will need to enter a queue with limited batchwise unstaking per epoch. Post-Shanghai, accumulated consensus layer rewards will become accessible and newly issued ETH rewards will be distributed automatically. For validators, there are two options to withdraw, one for partial withdrawals (PW) unlocking accrued staking rewards in excess of 32 ETH and one for full withdrawals (FW) exiting the entire balance including underlying principal and accrued rewards. For PW, validators must set their withdrawal credentials to enter the withdrawal queue. The withdrawal queue automatically unstakes rewards at 512 validators per epoch or 16 per slot (an epoch consists of 32 slots that last 12 seconds each and offer the opportunity to propose a block). For FW, validators need to enter an exit queue whose length is determined by the churn limit derived from the number of validators and a withdrawal queue. The exit queue is designed to keep stake and hence security volatility in check. At a churn limit of 7, it is limited to 50.4k ETH daily while the withdrawals queue is based on a round robin mechanism that currently takes around 4.3 days for the whole validator set. FW first pass through the exit queue, then through the withdrawal queue, PW only through the withdrawal queue. For FW, there is also a delay period of 256 epochs (27 hours) for non-slashed validators aiming to exit and 8’192 epochs (36 days) for slashed validators.
Beyond the activation of staking withdrawals, the next major upgrade for Ethereum dubbed Cancun will be centered around activating the Surge related Proto-Danksharding (EIP-4844), that is expected to massively boost L2 scalability. Initially planned for implementation with Shanghai, it was dropped along with EVM Object Format (EOF) and Simple Serialize (SSZ) to the next upgrade to avoid any delay for staking withdrawals.
Other EIPs included in the Shanghai upgrade are EIP-3651, EIP-3855, EIP-3860, EIP-6049.
Potential price implications
Shipping the Merge heavily fed into Ethereum’s value proposition and served as a confidence catalyst for further implementations of its ambitious roadmap. In 2023, we expect a vortex of upgrades, with the Shanghai and Capella upgrade initially enabling staking withdrawals. Heading into Q2 2023, a major question for investors is the potential price impact of unlocked ETH entering the market. Below, we analyze and project various sell scenarios based on a range of assumptions.
Activated withdrawals might induce a sustained period of increased yet limited liquid supply entering the market. To project the expected sell pressure, it is key to identify and understand the various staking entities, their cost basis and the withdrawal mechanics. As of writing, the majority of staked ETH is split across a few liquid staking providers such as Lido, and centralized exchanges such as Coinbase, Kraken, or Binance, see Illustration 1. Almost 25% of the market is held by unidentified stakers that are primarily made up of solo stakers or smaller groups running private staking pools. As all significant pools have otherwise been identified, the vast majority in that group is estimated to be independent individuals running their own validator nodes.
Illustration 1: Market share of liquid and non-liquid Beacon chain depositors
As of writing, 525k validators secure the Ethereum network with more than 16.8m ETH worth $27.5b and accrued 1.05m ETH in staking rewards at an average reward rate of 6.6%. Notably, around 50.2% of all ETH deposits is already liquid via liquid staking options from decentralized or centralized providers. That means that up to 75% of the overall stake is either liquid or solo staked, both indicating negligible sell pressure. In Table 1, we outline various scenarios from conservative to progressive selling taking the staking entity characteristics into account. In the below projection, we assumed that all withdrawal credentials are set at the time of the Shanghai upgrade and that all entities aiming to exit, be it partially or fully, do that immediately with activation of withdrawals via EIP-4895. We are not deriving potential sell pressure by accounting for the date of staking and the respective cost-basis. For the calculations, we round the numbers down to 500’000 validators and 1’000’000 ETH in accrued consensus layer rewards. It is important to note that depending on varying expectations, the amount of sell pressure from unlocked ETH shortly after the Shanghai activation can be calculated differently.
Table 1: Expected sell pressure scenarios
Overall, we expect negligible sell pressure from liquid staking providers in PW and FW but accounted for some due to depegs and liquidity issues in Liquid Staking Tokens (LSTs). LSTs however are reasonably pegged to the price of ETH already and we expect the delta to fully close as staking withdrawals get enabled. Most sell pressure will likely stem from non-liquid entities. Staking pools will probably set up new validators with their PW instead of selling. Solo stakers are usually strong long-term holders who are invested into their node setup and as such, we expect little sell pressure there. Illustration 2 indicates the expected sell pressure for the outlined scenarios.
Illustration 2: Expected daily sell pressure over time
We expect elevated sell pressure especially in the first couple days post-Shanghai as PW and FW coincide. However, even at the most progressive levels, the expected daily sell pressure does only account for less than 2% of the daily YTD spot volume sitting at roughly ~$8.2b (futures volumes are more than double that amount).
As PW calm down, the expected daily sell pressure induced by FW will likely not surpass 0.3% of the average daily spot volume and will arguably even be lower than the pre-Merge daily issuance of 13’509 ETH. A substantial amount of stake that enters the exit queue will be reshuffled or restaked to address necessary changes in infrastructure or activist staking to improve decentralization. Therefore, the overall share of exiting validators aiming to sell is limited. The exit queue will potentially be clogged for a sustained period as a large number of staking withdrawals will occur due to that recycling of validators and validators aiming to sell their stake. Moreover, larger entities with outdated structures, like Lido of which some validators are part of a multi-sig, will exit to optimize their setup. For instance, 10% of the validator set would currently take 31 days, 50% would take 158 days. As a result of the recent SEC clamp down on centralized staking providers, Kraken alone will already account for 7.3% that leaves the validator set. Most of this stake will most likely be restaked elsewhere however and not be sold. Despite the SEC being openly hostile towards crypto by containing it and limiting access, their actions will not only enhance regulatory clarity and protect clients from opaque staking services, but will arguably boost decentralized solutions. As such, Shanghai might shake up and even enhance the staking leaderboard towards a more distributed structure. There are many variables to consider while trying to assess the possible price implications. Therefore, one should take the number crunching with a grain of salt. Moreover, the activity around the Shanghai upgrade should not be observed in isolation.
Another important aspect projecting the potential selling pressure is the underlying cost basis at the time of the Beacon chain deposit. Only 20.5% of validators currently have a lower cost basis which means 79.5% of the stake has little financial incentive to sell now, see Illustration 3. While a substantial amount of early and non-liquid validators have a very low cost basis and potential unrealized profits, they are also the validators that had the highest conviction of being brave enough to enter the validator set very early on. As such, these Ethereum believers are also rather unlikely to sell. This again underpins the likelihood of only moderate to insignificant sell pressure from PW and FW. On the flipside, some sell pressure might be induced by validators being underwater for tax loss harvesting purposes as the deadline for tax returns in the U.S. expires in April 2023.
Illustration 3: Cost basis at date of Beacon chain deposit compared to current
While deposit activity shows steady growth and revitalized sentiment towards the upcoming Shanghai upgrade, Ethereum’s staking ratio remains polarizing low. Two years after the Beacon chain launch, only 14% of the ETH supply is staked while other smart contract platforms show ~40% and higher ratios, see Illustration 4. Notably, despite the lowest staking ratio, Ethereum dominates in absolute $ terms by a substantial margin ($29b compared to $9.1b of the second in staking market cap, Cardano). Until now, staking ETH required a risk profile willing to lock a risk-on asset for an indefinite period. As Shanghai de-risks staking and transforms ETH into a liquid commodity bearing staking rewards, we expect substantial inflows from investors and institutions that were hesitant before and avoided undefined lock-up periods. Hence, sidelined entities will finally be able to enter the stake and thereby lower the free-floating supply and even create demand. Moreover, new staking solutions like Liquid Collective will enable enterprise-grade staking for institutions. A higher staking ratio will also increase base layer security.
Illustration 4: Staking ratios of relevant smart contract platforms
Despite staking withdrawals being inactive and the persisting tough market conditions, the deposit contract saw back-to-back stable growth in Q1 ’22 (+25%), Q2 ’22 (+17.8%), Q3 ’22 (+8.2%) and notable growth in Q4 ’22 (+12.5%) driven by anticipation of Shanghai. The ETH balance in the deposit contract overall went up from 8.8m to 16.8m. The Merge was already a massive confidence boost that will only be reinforced by the Shanghai upgrade. Innovations like Distributed Validator Technology (DVT) and EigenLayer, that will boost the staking APR and utility via restaking, are bound to make staking even more attractive. DVT will lower slashing risk, infrastructure cost, stake centralization, single points of failure and ETH requirements to stake. The post-Merge supply dynamics will arguably tilt the inflation adjusted staking rewards even more attractive as it managed to be net deflationary in a market of comparably low activity. Heading into an uptrend that is known for stimulating on-chain activity, the sell pressure might decrease even more on the supply side and further boost the staking APR via e.g., MEV. As with Bitcoin halvings, we expect to see the new supply dynamics heavily in play as soon as blockspace demand hits elevated levels again. Ethereum’s inflation adjusted staking APR is already considered best in class and will likely have material impact across multiple industries. Introducing staking rewards to the largest smart contract platform will furthermore broaden institutional interest. As staking becomes less risky and liquidity ramps up, Ethereum's staking APR might become crypto's risk free rate and even yield curve products might be built around it. It is possible that staking also becomes crypto’s savings account making ETH more desirable.
We expect that the staking ratio will not catch up to Tezos- or Cardano-like levels as it is intrinsically capped . We expect the ratio to remain lower than the industry average as ETH inherits greater utility, yields lower staking APR with an increasing number of validators and is currently net deflationary. First, with 678 deployed protocols and highest DeFi TVL, the highest volume and ETH-denominated NFT market, consistently high blockspace demand, most DeFi innovations and its utility as gas token to access L2 ecosystems, ETH plays in another ballpark than alternative smart contract platforms. This range of utility will organically cap the expected staking ratio. Second, as more stake joins the Beacon chain, the staking APR will adjust lower as Ethereum’s issuance is inversely correlated to the validator set, see Illustration 5.
Illustration 5: Inverse relationship between validator set and staking APR
A declining staking APR might make staking less attractive to some validators and find equilibrium at a given level as staking rewards compete with yield sources available in DeFi or TradFi. And third, its post-Merge supply dynamics do allow non-staking entities to comfortably sit out of validating without being massively diluted by stakers. For instance, a PoS blockchain with very high inflation does pseudo force holders to stake to bypass dilution and hence induce a high staking ratio; Ethereum does not as it is currently deflationary. In the foreseeable future, one can expect a reasonable staking ratio between 40%-50%, with the corresponding consensus layer rewards then at 2.4% and 2.15% APR, respectively. Considering Tips and MEV, a validator would then still obtain an APR between 4-5% instead of currently 6%.
Conclusion and Outlook
Withdrawals will de-risk staking, increase capital efficiency of ETH that sat idle previously, and potentially transform ETH into a liquid commodity that might feature the risk-free rate of the decentralized and permissionless financial ecosystem. After the Merge, the Shanghai upgrade with the unlock of staked funds is expected to go hand in hand with a substantial confidence boost and hence attract new stake that was formerly hesitant. This hesitancy is very much indicated by a comparably low staking ratio that will likely grow significantly in the months following Shanghai. According to sustained growth since the Beacon chain launch, we expect that deposits outpace withdrawals in the medium- to long-term and that withdrawals will rather act as a structurally bullish unlock. A higher staking ratio will improve network security and reduce ETH’s free-floating supply that will arguably impact its value. Withdrawals will moreover allow activist staking, where stakers are free to actively reshuffle their funds to achieve more distributed validator pools. The obtained freedom will eliminate validators being trapped in regulatory crossfire. The recent SEC activities might moreover boost decentralized staking solutions and protect clients from opaque service providers by improving centralized staking solutions.
In the short term, we project elevated sell-pressure especially in the first week following Shanghai based on the characteristics of various validator entities and partial withdrawals coinciding with full withdrawals. We also expect more volatility in that period as well as in the days approaching Shanghai that might offer opportunities for investors. Following up, full withdrawals will arguably induce almost negligible sell pressure in the range of Ethereum’s previous daily PoW issuance. Overall moderate sell pressure induced by withdrawals is implied by several factors, such as the amount of stake that is already liquid, long-term Ethereum believers that run solo validators, the cost basis of the majority of deposits, as well as the attractive inflation-adjusted staking rewards.
Disclosure: at time of writing, the author holds ETH.