Data Analysis on ETH Selling Pressure After Shanghai Upgrade: How Many People Will Withdraw?

Original Title: 《Partial withdrawals after the Shanghai fork》

Original Author: Tripoli

Compiled by: 0x11, Foresight News

The specific date for the Ethereum Shanghai upgrade has yet to be determined, but it is approaching.

Discussions around selling pressure are once again prevalent, but it seems no one has deeply explored this topic from a data perspective. My earlier exploration of this topic's model is outdated but still being referenced; this article will serve as an update.

To read this article, you need to have a basic understanding of the mechanisms behind Ethereum withdrawals. If you are not familiar with the two types of credentials discussed below, they will be detailed later.

Compared to my earlier model, the main changes are:

• The control flow for switching between the two types of withdrawal credentials has been clarified.
• All withdrawals require the 0x01 credential, which is not set by early validators;
• The operation rate for adding the 0x01 withdrawal credential is limited to 16 times per block;
• All 0x01 credential validators will automatically perform partial withdrawals, processing up to 16 partial withdrawals per block.
• The process for full withdrawals remains unchanged, but 0x00 validators must set their 0x01 credentials before processing withdrawals.

For more information on this process, I recommend the following resources:

• ETH Withdrawal FAQ
• Christine Kim's detailed record

Which validators set which credentials?

Most analysts assume that the two types of validators (0x00, 0x01) have consistent characteristics, but this is a significant error. When we look at the deposits on the beacon chain in the figure below, we see that all genesis validators were created without setting the 0x01 credential, which severely distorts the data.

To date, about 20% of new validators have not set the 0x01 credential; they may be individuals using outdated guides and institutions with unupdated programs.

The primary adopters of the 0x01 credential system are Lido; 88% of liquid staking providers have set withdrawal credentials and control over 60% of validators, which are set to receive automatic withdrawals. Naive observers may worry about the centralization issues this could cause, but in this case, Lido's widespread adoption of the standard has a positive impact on the network.

Since Lido will recycle their rewards back to new validators to optimize their staking derivatives yield, the large adoption will block the partial withdrawal queue and reduce the initial volatility of withdrawals.

In addition to the heavy adopters of the 0x01 credential, most of Lido's validators were activated early, with the largest increase occurring in the spring of 2022 when their staking derivatives began trading at a discount. Therefore, the average age of Lido 0x01 validators is older than that of 0x01 validators and holds a significant amount of rewards that can be automatically withdrawn.

By dividing the accumulated rewards by the number of validators, we get the results shown in the figure below, where validators with Lido 0x01 certificates have averaged 1.23 ETH in rewards, while non-Lido 0x01 validators have averaged only 1.06 ETH.

Validators with 0x00 credentials are severely distorted by the genesis validators that have accumulated rewards for over two years. As of February 4, 2023, such validators have averaged 2.47 ETH in rewards (which can be withdrawn partially), while the best-performing validators have received a base reward of 4.9 ETH.

After the Shanghai Upgrade

Once withdrawals are enabled, Ethereum will begin processing partial withdrawals; scanning validators and distributing rewards to validators in the 0x01 pool. At the same time, 0x00 pool validators requesting to switch their credentials to 0x01 will begin filling the 0x01 pool. Both operations will occur at a rate of 16 validators per block.

The 0x01 pool will begin to heavily favor Lido validators. If the Shanghai fork occurs on January 29, 2023, the expected distribution of validators in the first block will be:

• 16 validators that initially set the 0x01 credential
• 9.95 Lido: a total of 13.19 ETH withdrawn.
• 6.05 non-Lido: a total of 6.89 ETH withdrawn.
• No validators that initially set the 0x00 credential.

By the 1,000th block (approximately 3 hours and 20 minutes later), the expectation will change to:

• 14.83 validators that initially set the 0x01 credential
• 9.22 Lido: a total of 12.22 ETH withdrawn
• 5.61 non-Lido: a total of 6.39 ETH withdrawn
• 1.17 validators that initially set the 0x00 credential: a total of 3.10 ETH withdrawn

By the 10,000th block (approximately 1 day and 9 hours later), the expectation will shift to:

• 7.50 validators that initially set the 0x01 credential
• 4.66 Lido: a total of 6.17 ETH withdrawn
• 2.84 non-Lido: a total of 3.23 ETH withdrawn
• 8.50 validators that initially set the 0x00 credential: a total of 22.57 ETH withdrawn

The peak position of the withdrawal rate (amount withdrawn per block) depends on the percentage of 0x00 validators that immediately switch their credentials, but in the most extreme case where all 0x00 validators request a change, the peak will be reached after 70 hours, with approximately 37.34 ETH withdrawn per block—of which 3.01 ETH comes from Lido validators.

Anyone suggesting that the peak will occur immediately has not deeply examined the data.

The animation below shows the evolution of daily partial withdrawals after different shares of 0x00 validators immediately switch their credentials. I believe the most likely scenario will fall between 25% and 50%, indicating that the peak rate will be reached earlier the next day and last for about two days.

In the most extreme case, the first day should see about 110,000 ETH in partial withdrawals (excluding Lido 0x01 validators). This figure sharply contrasts with those naive models that predict over 400,000 ETH in partial withdrawals on the first day.

Considering that non-Lido validators include most centralized institutions and liquid staking derivatives, it is reasonable to expect these validators to only sell a portion of their rewards. From this perspective, self-selected partial withdrawals seem unlikely to lead to catastrophic events in the market. The decision for automatic partial withdrawals may reduce the initial selling pressure by blocking the queue with validators who do not intend to sell.

Before the Shanghai Upgrade

Data is not a definitive outcome, so a natural question to ask is: how should we expect the profiles to change before withdrawals are enabled? My intuition is as follows:

All types of validators will continue to earn rewards, and new validators will keep emerging.

Compared to the existing 310,000+ validators, new 0x00 validators will be very scarce, so their average balance will continue to rise, reaching 2.5 to 2.6 ETH per validator.

The number of Lido 0x01 validators will continue to increase, while the average balance per validator will also rise to 1.3 to 1.4 ETH.

The number of non-Lido 0x01 validators will accelerate, putting downward pressure on the average balance, which I expect to be in the range of 1.0 to 1.1 ETH per validator.

The ultimate result of these increases should be that the maximum withdrawal rate increases by 10% compared to the estimates in this article.

Increasing the number of validators (especially 0x01) before enabling withdrawals will flatten the curve; extend the duration of selling pressure but slightly reduce the peak of selling pressure.

Full Withdrawals

The process for full withdrawals remains unchanged. You can track the outflow limits in real-time here, but the outflow limit factor at the start of withdrawals may be 8, indicating that 57,600 ETH can be withdrawn daily through full withdrawals.

A queue will emerge to activate new validators and for validator exits. The double-ended queue will lock the outflow limit and effective validator count at a value until one side of the queue diminishes. I suspect the community will view the Shanghai upgrade as a sufficient de-risking event, leading to a surge in new validator deposits, but there will also be many validators who believe staking is not the right choice.

Over time, the number of validators may increase, but anything can happen in the short term.