.svg){kind=icon}
Now hiring!
•
There are concerns about large validator exits occurring shortly after the Shapella upgrade. In this article, we examine whether these fears are justified and attempt to estimate how many validators are expected to exit and how many will join post-upgrade.
First, we explain the distinction between partial and complete withdrawals, noting that not all validators can unstake immediately after withdrawals are enabled. We then examine the various drivers of withdrawal and staking demand, such as:
The Shapella upgrade will introduce two distinct types of withdrawals: full (also known as exits) and partial (staking reward collection). For more detailed information on the withdrawal process, we refer you to our recent article in collaboration with Twinstake.
This section will focus on the expected ETH outflow associated with full and partial withdrawals.
Complete withdrawals allow validators to withdraw their entire stake after going through an exit queue and a withdrawal period. This process was designed to prevent sudden changes in the number of validators, ensuring network security.
How Does The Exit Queue Work?
Full withdrawals must go through both the exit queue and the withdrawal period to be considered withdrawable. The exit queue is the main bottleneck and works in a first-in, first-out (FIFO) fashion, limiting the number of validators who can withdraw in each epoch (period of 32 slots) to a dynamic churn limit. The following equation determines the limit:
The minimum churn limit is currently set to four, meaning that at least four validators must be able to withdraw every epoch. The graph below shows the relationship between the number of active validators in the network and the churn limit.
.png)
Given that there are close to 530,000 validators as of February 27th, 2023, the churn limit is 530,000 / 65,536 = 8.08, which means that only eight validators are allowed to exit per epoch (6.4 minutes). Once the number of validators exceeds 589,824 (65,536*89), nine validators will be allowed to exit.
The chart below shows the days for a given proportion of the validator set to exit the queue altogether. If 10% of validators decided to unstake simultaneously, they would have to wait in the queue for 16 days on average.
.png)
Once validators go through the exit queue, they must wait for the withdrawal period to end. The withdrawal period will be 256 epochs (~27 hours) for validators who have not been slashed and 8,192 epochs (~36 days) for those who have. Once the withdrawal period is over, the validators will wait for their withdrawal to be processed (potentially up to 5 days).
Slashed actors have a longer withdrawal period due to the slashing penalty mechanism. Once slashing is detected, the minimum penalty is issued, which has a fixed size. After 18 days, a midterm attack multiplier penalty is applied, proportional to the number of other slashings in approximately the previous 36 days.
Please refer to our previous article for more information on the exit queue.
Only 40% of Validators Will be Eligible to Withdraw From Day One
It is also important to note that to be able to withdraw their stake, validators will need a withdrawal address of the 0x01 format. The option of setting a 0x01 withdrawal address when joining the validator set was only introduced in March 2021. However, the validators that joined earlier were stuck with the old 0x00 format addresses without the option of switching to the new format. Currently around ~310,000 or ~58% of Ethereum validators have the old 0x00 format withdrawal addresses. The Shapella upgrade will enable those with the old format withdrawal addresses to switch to the eligible 0x01 addresses. However, the rate of withdrawal address switching will be limited to 16 per block, meaning it would take around three days for all 0x00 addresses to change to 0x01 and become eligible to withdraw their ETH.
After the Shapella upgrade, accumulated staking rewards (balances over 32 ETH) will be withdrawn automatically if validators have the required withdrawal credentials. This is referred to as partial withdrawals. Partial withdrawals are processed according to the indices assigned to individual validators when joining the Beacon Chain (as opposed to first-in, first-out), with a withdrawal pointer looping through these indices and withdrawing at a rate of 16 validators per block. As a result, it would take 4.7 days for the pointer to go through all the validator indices, assuming they are all eligible. However, since only 42% of validators have the required credentials, partial withdrawals will be processed within approximately 2 days.
Having covered the withdrawal process and eligibility criteria for validators, we can delve into the factors that drive demand for staking withdrawals and activations. Several factors may influence the relationship between the two, including:
It is commonly believed that early validators will be especially eager to use the Shapella upgrade to unstake. Currently, a significant portion of active validators, ~56.9%, have been staking for over a year, so if the aforementioned belief holds true, the majority of the validator set may wish to exit once the upgrade is live. To test this hypothesis, we analyzed withdrawals on Polygon and drew comparisons.
.png)
Insights From Polygon Staking
Insights from Polygon staking contradict the common belief that validators who have staked for a prolonged period are more likely to withdraw.
Withdrawals on Polygon have been enabled since the inception of the staking contract in 2020. According to Dune Analytics, 36,101 wallets have staked MATIC on Polygon. Of those, 55% have only withdrawn rewards and not their entire stake. For those who did unstake, most did so within the first three months. The median staking time was 92 days.
Small-stake validators on Polygon are less likely to exit, with only 42% of them having done so. Large MATIC stakers, holding 10 million or more tokens, are also relatively unlikely to exit. However, we note that middle-sized validators are more likely to unstake. This suggests that large Ethereum holders will likely retain their staked ETH, while those with smaller holdings may be more inclined to exit.
Based on these findings, we do not anticipate validators that have been staking ETH for over a year to exit, especially if they are associated with entities that control many validators.
The chart below provides a breakdown of staking deposits by user type. CEXs (note that Coinbase’s cbETH is classified as a CEX and not liquid staking) make up 28% of the total ETH staked, but they are a volatile source. CEXs may be more likely to withdraw their staked ETH since Kraken was recently fined by the SEC for the illegal sale of securities. Kraken, in particular, will unwind its US staking operations after the Shapella upgrade. Based on the SEC filing, we estimate that between 25% to 30% of Kraken’s staked ETH is from US investors, which represents between 300k to 400k ETH (2% of total staked ETH) that will be unstaked.
.png)
It is yet to be seen whether Kraken will exit its validators all at once or gradually over several weeks. Regardless of the approach, the withdrawals are expected to be manageable and not cause significant congestion to the exit queue. Nonetheless, monitoring other SEC actions against US CEXs in the coming weeks is essential, as they could destabilize the staking market and lead to large exit queues.
Crypto price volatility is often seen as a critical factor in redemptions. Analysis of MATIC redemptions reveals a non-linear correlation between MATIC volatility and the amount of unstaking. The probability of a large unstaking of MATIC over 10 days (i.e. more than 2% of total staked MATIC) increases significantly when MATIC has a 15%+ drawdown. Historically, when MATIC has dropped more than 15% over 10 days, there has been a 41.8% probability of a large unstake vs 4.3% otherwise.
We observed a similar pattern with Solana staking as the main unstaking waves occurred during the 2021 flash crash and FTX collapse.
Inversely, sustained token price appreciation does not lead to increased staking. The amount of MATIC staked nearly doubled from 2022, despite the price of MATIC decreasing by a factor 3. Similarly, the amount of ETH staked increased moderately between July and December 2021, while the price multiplied by 2.5x.
.png)
.png)
According to beaconcha.in, there are ~530,000 active validators, with an average balance of 34 ETH. Since anything over 32 ETH can be considered staking rewards and will be automatically withdrawn through partial withdrawals, approximately 1,060,000 ETH or 6.2% of the total staked ETH will be available for partial withdrawals.
However, as we saw earlier, most validators have yet to switch to 0x01 addresses. Assuming all remaining validators switch credentials once the Shapella upgrade is live, the highest volumes of partial withdrawals will only occur two days after the upgrade. As a result, most of the staking rewards will take two to three days to be withdrawn.
Since validators don’t receive any additional yield for staking more than 32 ETH, a large portion of the partially withdrawn ETH will likely be restaked in new 32 ETH units that generate extra yield. The queue mechanics for activations (becoming a validator) are similar to those for withdrawals — only a limited number of validators can join the validator set each epoch. For example, if all 1,060,000 ETH are restaked, a validator could take up to 20 days to join the network.
While waiting in the activation queue, validators don’t generate any staking rewards. This may lead to a more significant premium for liquid staking derivatives, which will continue to generate yield while ETH in the activation queue doesn’t. Based on 5% staking returns, a 20-day activation wait would warrant a 27 basis points premium (5%/365*20). The activation queue may be longer if the Shapella upgrade encourages ETH holders to stake more.
The primary demand for staking will come from the idle ETH in wallets. Staking yields are close to 5%, meaning that not staking incurs a high opportunity cost. Therefore, holders have a solid incentive to stake their idle ETH, provided gas fees are reasonable, and they have no security or withdrawal worries. Below we examine the primary uses of the top 1,000 ETH holders. These holders, including the Beacon Chain, control 62.54% of the total ETH in circulation.
Aside from staking, a significant portion of ETH is held by individual wallets. After analyzing transactions, we can divide these wallets into three distinct groups:
The first two groups of wallets, those with low turnover or those accumulating ETH, likely have little need to spend it in the short term. Therefore, they would benefit from staking their ETH to earn interest. Together, these two categories represent 21.58% of total ETH holdings, which is higher than the amount currently staked.
In addition to these categories, treasuries are prime targets for staking, as they collectively own over 1% of the total ETH supply. Examples of DAOs in this category include ENS, BitDAO, GuildFi, and Lido, which have amassed a war chest and plan to spend it over time. These treasuries could stake the majority of their ETH to improve their treasury management.
CEXs account for approximately 10% of the total circulating ETH, while DeFi protocols account for around 5%. We anticipate that ETH will remain on those platforms, as the need to stake ETH is less pressing than for idle wallets.
We have observed that the Top 1000 ETH holders are likely to generate significant staking demand. Although we have not yet analyzed the remaining ETH wallets (which account for 37% of the total ETH), we anticipate that they, too, will contribute to the demand for staking. Smaller wallets may opt for liquid staking as a more cost-effective solution.
Although predicting the exact number of withdrawals post-Shapella is challenging, we expect only a limited number to exit immediately after withdrawals are enabled, as most validators will not have the necessary withdrawal credentials. The week following the Shapella upgrade should see substantial partial withdrawals, providing a good cushion for validators needing liquidity without requiring them to fully withdraw . Some partial withdrawals will be reinvested, but it will take some time for all the new validators to join the network and earn rewards due to the effect of the activation queue. We also anticipate increased demand from long-term ETH holders with idle ETH in their wallets. This could cause liquid staking coins, such as stETH, to trade at a premium compared to their fair value. For long term projections, past staking activity on Polygon and Solana suggests that validator exits are usually relatively small, barring a significant liquidity event. Nevertheless, the SEC’s recent crackdown on Kraken, which will force it to unwind its US staking operations, creates major uncertainty in the industry. Therefore, it is necessary to monitor how the SEC’s actions develop and whether they expand their reach to other cryptocurrency exchanges (CEXs), forcing more withdrawals to follow.
This article has been prepared for the general information and understanding of the readers. No representation or warranty, express or implied, is given by Nethermind as to the accuracy or completeness of the information or opinions contained in the above article. The views and strategies described may not be suitable for all readers. Both past performance and yield may not be a reliable guide to future performance. All case studies are shown for illustrative purposes only and should not be relied upon as advice or interpreted as a recommendation. The article is based on the scope of materials and documentation publicly available to Nethermind, and it does not indicate Nethermind’s endorsement of any particular asset, project or team, nor guarantee its security. No third party should rely on this article in any way, including for the purpose of making any decisions to buy or sell a product, service or any other asset. To the fullest extent permitted by law, Nethermind disclaims any liability in connection with this article, its content, and any related services and products and your use thereof, including, without limitation, the implied warranties of merchantability, fitness for a particular purpose, and non-infringement. Nethermind does not warrant, endorse, guarantee, or assume responsibility for any product or service advertised or offered by a third party through the product, any open source or third-party software, code, libraries, materials, or information linked to, called by, referenced by or accessible through the article, its content, and the related services and products, any hyperlinked websites, any websites or mobile applications appearing on any advertising, and Nethermind will not be a party to or in any way be responsible for monitoring any transaction between you and any third-party. You should use your best judgment and exercise caution where appropriate. FOR AVOIDANCE OF DOUBT, THE ARTICLE, ITS CONTENT, ACCESS, AND/OR USAGE THEREOF, INCLUDING ANY ASSOCIATED SERVICES OR MATERIALS, SHALL NOT BE CONSIDERED OR RELIED UPON AS ANY FORM OF FINANCIAL, INVESTMENT, TAX, LEGAL, REGULATORY, OR OTHER ADVICE.
.svg)
The simplest solutions to the hardest
problems in blockchain
Nethermind © 2024
.svg)
.svg)
.svg)
.svg)
.svg)
.svg){kind=icon}
.svg)
.svg)
.svg)
.svg)
.svg)
.svg)
.svg)
.svg)
.png)
.png)
%20(1).png)
.png)
