Data on Ethereum 2023: pledge amount +60%, annual supply -0.28%, average TVL of 12 L2 +333%

This article analyzes the data of liquidity staking and L2 in 2023 to review and look forward to the development of Ethereum.

Written by Carol, PANews

In 2023, Ethereum has two main story lines.

The first line relates to liquidity staking. In April, Ethereum completed the Shanghai upgrade, officially opened pledge withdrawals, and entered the era of "interest rate control". However, contrary to earlier market concerns, the upgrade of Ethereum not only did not cause large-scale selling pressure, but instead attracted more deposits as the currency price rose, making LSD the most popular DeFi track.

The second line is related to Layer2 (hereinafter referred to as L2). Arbitrum’s airdrop not only renewed the “myth of wealth creation” once again, but also raised expectations for other L2 airdrops. In addition, with Base's instant success and the Cancun upgrade approaching, technical keywords such as modularization, parallel EVM, data availability (DA), and decentralized sorters are becoming ubiquitous. The rapid development of L2 and the rapid rise in prices of representative assets such as OP have made L2 the most anticipated field in 2024.

PAData, a data news column under PANews, analyzed liquidity staking and L2 data in 2023 to review and look forward to the development of Ethereum:

• After Shanghai's upgrade to the end of the year, the total amount of Ethereum deposits was approximately 17.3061 million ETH, and the total redemption amount was approximately 8.1135 million ETH. The cumulative total pledged amount increased by 10.8242 million ETH, an increase of approximately 60%, but the pledge growth rate dropped rapidly. This means that people are no longer as keen on pledging their deposits as they used to be.
• Will Ethereum’s pledged deposits squeeze out other DeFi activities? The correlation analysis results show that within a small cycle, when the price clearly drops, more funds flow from DeFi to pledge deposits, and Ethereum pledge activities may have a "siphoning" effect; when the price clearly rises, funds simultaneously flow from pledge activities There is no reverse "siphon" outflow from DeFi activities.
• In 2023, the total amount of fees burned by Ethereum is approximately 1.0935 million ETH. The supply decreased by approximately 341,800 ETH, a decrease of approximately 0.28%. ETH has generally experienced slight deflation, but the magnitude and trend of deflation are not significant.
• Based on the basic situation of 34 L2s, 11 of them each use Optimistic Rollup and ZK Rollup. And most of them are in the early stages of technology, and 17 are in the STAGE 0 stage that only implements the status submission function. Including the popular OP Mainnet, Base, zkSync Era, Starknet, etc.
• The highest TVL in L2 is Arbitrum One, reaching US$9.370 billion, followed by OP Mainnet, reaching US$6.050 billion. Companies with higher TVL growth during the year include Manta Pacific, Starknet, and Polygon zk EVM.
• The largest cumulative number of transactions in 2023 was Arbitrum One, with more than 275 million transactions. Followed by zkSync Era, with more than 218 million times. The TPS of the two once exceeded Ethereum at the end of the year.
• In 2023, the number of independent addresses and daily active addresses of L2 will increase significantly. The highest annual growth rate of independent addresses was Base, with an increase of over 7166%. Mantle has the highest increase in daily active addresses over the past few years, with an increase of over 886%.
• Among the four major Rollup L2s, zkSync Era has the highest annual revenue, reaching US$22.26 million, followed by Arbitrum, reaching US$16.53 million. Both Base and Optimism are less than $6 million.
• The currency price with the highest increase in 2023 is IMX, which has increased by more than 454%, followed by METIS and OP, which have both increased by more than 300%. ARB's price performance was mediocre, rising only 14.81% throughout the year.

01. After Shanghai’s upgrade, the total pledged amount increased by nearly 60%, but the growth rate declined. The total amount burned throughout the year exceeded 1.09 million ETH, achieving deflation.

From the successful completion of the Shanghai upgrade on April 12 to the end of the year, the total deposits of Ethereum were approximately 17.3061 million ETH, and the total redemption amount (including Principal and Rewards redemptions) was approximately 8.1135 million ETH. Among them, the total redemption amount showed an inverted "U" shape, with higher levels in April and December, exceeding 1.96 million ETH and 1.58 million ETH respectively. The total deposits are basically stable. Except for May, which attracted 3.82 million ETH, the other monthly total deposits are about 1.68 million ETH.

After the upgrade in Shanghai, the cumulative total pledged amount during the year increased from 18.1659 million ETH to 28.9901 million ETH, an increase of 10.8242 million ETH, an increase of approximately 60%.

From the perspective of staking entities, Lido and Coinbase are the two entities with the largest increase in total pledges in 2023, with increases of 4.5597 million ETH and 1.7087 million respectively, while the increments of other entities are no more than one million ETH. Among them, the total pledged amount of Figment, Kiln, OKX and Rocket Pool increased by more than 500,000 ETH, which is a significant increase.

However, judging from the full-year data, the growth rate of the total amount of Ethereum pledges is declining rapidly. In May, the average daily month-on-month growth rate of the cumulative total pledged amount reached a peak of 0.48%. However, since then, the value has declined month by month. By December, it had dropped to 0.03%. The downward trend is very obvious. This means that people are no longer as keen on pledging their deposits as they used to be.

According to data from DuneAnalytics, although the staking interest rate of Ethereum has dropped from 4.18% at the beginning of the year to 3.09% at the end of the year, it is still a stable financial income. There has always been a voice that believes that Ethereum’s high pledge interest rate may attract more deposits, thereby squeezing other on-chain activities and affecting the long-term development of the Ethereum ecosystem. So judging from the data performance, is this assumption tenable?

PAData has calculated the correlation between the daily changes in the cumulative pledge total and the daily changes in TVL after Shanghai's upgrade, and also the correlation between the daily changes in the total redemption amount and the daily changes in TVL. Here, TVL uses the ETH pricing method, and according to the definition of DefiLlama, TVL does not include the lock-up amount of the liquidity staking agreement.

If there is a negative correlation (≤-0.5) between the daily change in the cumulative pledge total and the daily change in TVL, or a positive correlation (≥0.5) between the daily change in the total redemption amount and the daily change in TVL, it will also That is, when a higher daily increase in the total amount of pledges and a lower daily increase in TVL occur at the same time, or when a higher daily increase in the total redemption amount and a higher daily increase in TVL occur at the same time, Ethereum staking activity can be considered There is a "siphon effect" and vice versa.

The correlation analysis results show that, in general, it cannot be proven that there is a "siphon effect" in Ethereum staking activities.

However, if the time period is shortened to weekly, then there may be a "siphon effect" on Ethereum staking activities during part of the time period. For example, in mid-to-early August (around August 6th to August 19th), the correlation coefficient between the daily change in the total redemption amount and the daily change in TVL was 0.671, and the daily change in the cumulative pledge total and the daily change in TVL were 0.671. The correlation coefficient between day-to-day changes was -0.533, when the weekly average price of ETH fell from $1,844 to $1,659. This shows that when the price clearly drops, more funds flow from DeFi to pledge deposits, and Ethereum pledge activities may have a "siphoning" effect.

Another typical time period occurs in mid-to-late October (around October 15th to October 28th). The correlation coefficient between the daily month-on-month changes in the cumulative pledge total and the daily month-on-month changes in TVL is 0.708. The correlation coefficient between the month-on-month change and the daily month-on-month change in TVL is -0.579, when the weekly average price of ETH rose from $1,583 to $1,765. This shows that when prices clearly rise, funds flow out of staking activities and DeFi activities at the same time, and there is no reverse "siphon".

One of the important prerequisite changes for the Shanghai upgrade is the implementation of EIP1559, which makes it possible for Ethereum to achieve deflation. In 2023, the total amount of fees burned by Ethereum is approximately 1.0935 million ETH. Among them, 204,600 ETH was burned in May, which was the month with the largest amount of burning in the whole year; 41,300 ETH was burned in October, which was the month with the least amount of burning in the whole year. At the end of the year, the burning volume rose again to over 100,000 ETH.

Under such circumstances, the supply of ETH in 2023 dropped from 120.53 million ETH to 120.18 million ETH, a decrease of approximately 341,800 ETH, a decrease of approximately 0.28%. In other words, ETH will generally achieve slight deflation in 2023.

Judging from the day-to-day changes, the supply in 2023 will decrease by 939 ETH per day on average. In terms of trends, the magnitude of day-to-day changes in supply is decreasing. In other words, it can be understood that the trend of deflation is very weak and not significant.

02. The average annual TVL growth of 12 L2s exceeded 333%, and the total revenue of zkSync Era was US$22.26 million.

2023 is a year of rapid development in the L2 field. Not only does OP Stack occupy most of the market, but also a variety of new technologies such as modular development, parallel EVM, decentralized sequencers, and third-party DA solutions have emerged. trend. The blueprint seems to be ready, so what is the actual development of L2?

Judging from the basic situation of 34 L2s counted by L2BEAT, 11 of them each use Optimistic Rollup and ZK Rollup, which are about the same number. Another 8 use Validium and 4 use Optimium. The main difference between different technical architectures is the choice and combination of different data availability (DA) and proof systems (proof).

Most of these L2s are general-purpose, and only a few are specific to Exchange or NFT, such as dYdX v3 and Immutable X.

Most of these L2s are in the early stages of technology. A total of 17 L2s are at STAGE 0, including the popular OP Mainnet, Base, zkSync Era, Starknet, etc. Only three L2s have reached STAGE 1, including Arbitrum One, dYdX v3 and zkSync Lite.

From a technical point of view, the main difference between STAGE 1 and STAGE 0 is that STAGE 1 implements state submission to L1 and partially implements the certification system, including that users can exit based on certain review procedures and can be more centralized than the Security Council. Participants will withdraw within 7 days after unnecessary upgrades, etc.

There are also 2 L2s that have reached the higher STAGE 2. The technical difference between them and STAGE 1 is mainly that STAGE 2 further improves the proof system, including fraud proof submissions that are only open to whitelist participants and provable vulnerabilities on the chain. Irrelevant upgrades provide less than 30 days of exit time, the Security Council's actions are not limited to provable vulnerabilities on the chain, etc.

Currently, the highest TVL (referring to the total value locked in the Ethereum custody contract, which may include external bridges and locally created assets) in L2 is Arbitrum One, which reached $9.37 billion, followed by OP Mainnet, which reached $6.05 billion. Except for these two, the TVL of other L2s does not exceed US$700 million. The market share of L2 is basically shared by Arbitrum One and OP Mainnet.

However, in 2023, there will still be many "new faces" in the L2 market besides the two heroes. Among the 12 L2s with the highest TVL, 6 will launch their mainnets only in 2023, accounting for half of them. These new L2s took advantage of the momentum and achieved high TVL growth during the year. For example, the TVL growth of Manta Pacific exceeded 5387%, and the TVL growth of Starknet and Polygon zk EVM both increased by over 2000%.

Others with higher TVL increases include Metis Andromeda and Linea, which exceeded 688% and 488% respectively. In addition, L2 that was launched earlier such as Arbitrum One, OP Mainnet, and Immutable X also achieved a TVL increase of more than 200%.

Due to data availability, the subsequent analysis of on-chain data is limited to some L2s with higher TVL.

In terms of the number of transactions, Arbitrum One has the highest cumulative number of transactions in 2023, with more than 275 million transactions. Followed by zkSync Era, with more than 218 million times. OP Mainnet also exceeded 136 million times, and the cumulative number of L2 transactions in other statistical ranges was less than 100 million times, and some were even less than 10 million times.

It is worth noting that due to the inscription craze, Arbitrum One and zkSync Era once surpassed Ethereum in TPS at the end of the year, withstanding the test of high-frequency interaction.

From the perspective of user scale, L2 within the statistical range has achieved very significant growth in 2023. The highest annual growth rate of independent addresses was Base, with an increase of over 7166%, followed by OP Mainnet, with an increase of over 3683%. The one with the highest increase in daily active addresses over the past several years was Mantle, with an increase of over 886%, followed by Metis Andromeda, with an increase of over 455%. Overall, this means that not only are more people starting to use L2, but they are using it more frequently.

Although L2's development is bright, for now, its funding scale and user scale are still relatively limited, which reflects in revenue that L2's income is not high. After statistics on the income of the four main Rollup L2 categories, it was found that zkSync Era had the highest annual income, reaching US$22.26 million, followed by Arbitrum, reaching US$16.53 million. Both Base and Optimism are less than $6 million.

Moreover, L2's revenue has not kept pace with its number of transactions, because an increase in the number of transactions usually means an increase in DA costs. For example, the high income of zkSync Era mainly comes from the second quarter, not the fourth quarter when Inscription is hot. How to balance fee income and DA costs may become an important issue that needs to be solved for the future development of L2.

If the rapid development of L2 is mapped to the currency price performance, then there are still some differentiations and dislocations. In L2 with high TVL, there are still not many tokens that have been issued. Among them, the currency price with the highest increase in 2023 is IMX, which has increased by more than 454%, followed by METIS and OP, which have both increased by more than 300%. However, despite ARB's other impressive data performance, ARB's price performance was mediocre, rising only 14.81% throughout the year, which was lower than MNT and MATIC.

More L2 projects may issue tokens in 2024. How to balance fee income and DA costs, and how to provide value support for tokens will still become important issues that need to be solved for the future development of L2.