A retrospective review of the historical development of Ethereum: including address count, on-chain transactions, and distribution of holdings

2020/2/28

share

How is the development of Ethereum? We seek answers from historical data: growth in the number of addresses, total on-chain transactions, holdings distribution, and DeFi development.

Original Title: "Viewpoint | Understanding the Development of Ethereum Through Historical Data"
Author: Joel John
Translation & Proofreading: Min Min & AJ

The goal of Ethereum is to become a world computer. Along the way, many things have happened: from being a cat salesman, to being attacked and forked, to fighting against authoritarian regimes, and even creating a new financial structure. There are many opinions about Ethereum - some say it's a currency, some say it's a failure, some say it's a mutable ledger, and so on. These evaluations are not entirely fair. Ethereum is like a knife, it can be used as a tool or a weapon, depending on the user. Therefore, I started to explore the development of Ethereum on my own. My interest in this topic was largely sparked by an article I wrote in May 2019 about decentralized finance. I found that almost all decentralized finance applications are built on Ethereum. This article is my attempt to organize all the data related to Ethereum.

Advertisement - Continue scrolling for more content

Image Source: https://twitter.com/joel_john95/status/1219543404117250049

In this article, I have avoided directly comparing Ethereum with Bitcoin. Just as the financial system can produce various commodities, currencies, and financial instruments, I believe the digital asset ecosystem can also undergo different experiments. If we believe in decentralization, one way to achieve decentralized mechanisms is to maintain multiple ledger backups, as a single ledger is more susceptible to black swan events. In a world dominated by "ledger supremacy," such experiments have no room to grow. But for the same reasons, I support multiple ledgers as long as they can fulfill their roles without error. I hope this article reflects the development of Ethereum rather than belittling Bitcoin or Ethereum.

Activation of Ethereum addresses (red line represents new addresses, black line represents active addresses, blue line represents total activated addresses)

There are a total of 69 million addresses on Ethereum, with only about 250,000 addresses active each day. Surprisingly, 50,000 of them are new addresses. These numbers may not seem significant due to the low cost of creating new wallets or transferring assets on Ethereum. What's interesting to note is that the number of active addresses and new wallets is much higher compared to the ICO boom in 2017. One key factor could be the switch of USDT to the ERC-20 standard and the rise of many mixers on Ethereum. In 2019, around 226,000 different wallet addresses interacted with decentralized finance applications. These users are likely power users who frequently use wallets. Although decentralized finance is attractive, the future of wallet activity on Ethereum may be supported by non-fungible tokens (NFTs) and digital gambling. Games like Axie Infinity and Gods Unchained are driving a new generation of users to use the Ethereum network unconsciously. This transition is similar to moving from Internet Relay Chat (IRC) to WhatsApp, where the former is functional, and the latter becomes mainstream. With projects simplifying ledger interactions, we may see more active addresses in the future.

Red represents the number of transfers, black line represents the number of transactions

If the number of active wallets and new wallets is used as an indicator of user interaction with Ethereum, then the number of transactions is a measure of their transaction frequency. There is a striking similarity between the two charts. Compared to the ICO frenzy year, transaction and transfer volumes have not seen a significant decline. This is largely attributed to the maturing of decentralized finance and liquidity pools. If the total number of addresses on Ethereum were to decrease, it could be inferred that the remaining users are power users who initiate many transactions. However, since both of these indicators have not decreased, it indicates that

1. Since 2017, new users have offset the number of lost users
2. The transaction frequency of new users is similar to that of previous ICO investors

The real significant change lies in the next indicator—the average transaction size on Ethereum.

Red line represents the daily average transaction size, black line represents the total daily transaction volume, both measured in USD

The average transaction size peaked on August 11, 2017, reaching nearly $40,000. The total transaction volume on the Ethereum blockchain reached its peak on January 14, 2018, at $33 billion. These two data points can be connected to the popular use cases of Ethereum at the time. The ICO boom first appeared in August 2017 and by the end of September had attracted over $1 billion in funding. On the other hand, after Bitcoin experienced a price correction in December 2017, the altcoin craze peaked in January 2018. However, it is crucial to look forward and not dwell on the past. Active addresses and transaction volumes show the number and frequency of transactions on the Ethereum ledger, while the average transaction size reflects the volume of funds transferred. As of the writing of this article, the average transaction size is around $138. Based on these indicators, the mass adoption of digital assets is evident.

In summary:

1. The number of transactions has increased sixfold since February 2017
2. The number of active addresses has increased by approximately tenfold since February 2017
3. The total on-chain transaction value has increased by about twelvefold (from $11 million to $1.28 billion)
4. However, the average transaction size has decreased by around 99.5%

As non-fungible tokens mature, the average transaction size may continue to decrease. Typically, Ethereum users do not need to transfer large amounts of Ether. An increase in the number of small traders indicates Ethereum's strong appeal to users. This can also be measured by the amount of Ether held in the top accounts.

Red line represents the amount of Ether held by the top 100 wallets of exchanges, black line represents the amount of Ether held by the top 100 wallets outside of exchanges

When considering the percentage of Ether held by the top 100 addresses, it becomes more apparent that Ethereum's mass adoption trend is underway. Including exchange wallets, the top 100 addresses hold about one-third of the total Ether supply, roughly 32%. This figure may seem alarming for a project supporting decentralized finance, but there is a decreasing trend. According to Glassnode data, Bitcoin exchanges hold about 11% of the circulating supply, while Ethereum exchanges hold about 8%. Since November 2018, Ethereum whales seem to be hoarding Ether. Several explanations for this include:

1. Investors who diversified their investments during the peak of the 2018 bull market are now buying Ether. They can now buy back the Ether they sold in previous years at lower prices.
2. Due to the significant drop in Ether's price since its all-time high, some investors are no longer selling Ether, causing the percentage to remain stagnant without a significant decrease.
3. As more retail traders buy Ether, the cost of holding a certain percentage of Ether increases. It is becoming more challenging to form new whales on the Ethereum network.

Another reason for these whales not selling more Ether is the ability to borrow and lend in USD terms without traditional financial infrastructure (decentralized finance). Analyzing decentralized finance through the Pareto principle is evident. Typically, 3 to 5 whales control 60% to 80% of total assets within a product. Additionally, Ethereum has retained a group of large holders who are bullish on the network's future. This can be seen as a positive sign of support.

Black line represents the total fee amount, red line represents the average fee size per transaction

The total transaction fees on the Ethereum network amount to approximately $242 million. Interestingly, the peak of transaction fees was not during the 2017 ICO boom but in 2018, when the total transaction fee amount was $160 million. In comparison, the transaction fees in 2019 were only around $34 million, similar to 2017. In February 2018, the cost per transaction surged to $5.50, but has since dropped to $0.10. Although transaction costs are currently very low, Ether's price has more than doubled from $0.03 before the price surge in February 2017. Fluctuating transaction fees can have a negative impact on products built on Ethereum. Developers must consider the transaction costs borne by small traders. Decentralized computation and storage services often face the risk of high gas costs, which may deter customers. Some projects are working to address this issue. For example, through Gasless by Mosendo, users can transfer DAI without holding Ether to pay gas fees.

Similarly, a new wave of API-layer projects will soon emerge, enabling end-users who do not hold Ether to transact while charging businesses directly. Nuo Network has already launched an early version of this. In the long run, if the price of Ether rises significantly, more complex derivatives and futures will likely emerge to hedge transaction fees on a large scale, similar to the traditional financial world today.

Black portion represents the block reward size, red line represents the fee size

When comparing the accumulated block rewards with the total accumulated transaction fees, an interesting feature of Ethereum is revealed. Over the past five years, the total block reward amount reached $150 million. Essentially, Ethereum has paid miners around $300 million so far, but compared to Bitcoin, it pales in comparison. Bitcoin has paid miners around $16 billion to date. Judging from the trading volume we see in the derivatives market, it is clear that due to the fee mechanisms of both, people prefer mining Bitcoin over Ethereum. A common explanation I found is that Bitcoin miners also consider mining Ethereum as a side job. Typically, miners tend to hold net long positions as they anticipate price increases. The fluctuation of miner balances with price volatility further supports this point.

Source: San Graphs (blue line represents ETH price, dark blue portion represents miner addresses holding Ether)

Several factors may have influenced miner behavior over the past few years. It seems that between January 2016 and January 2017, due to the lower transaction prices of Ether, Ethereum appeared to be a popular experimental project within the geek community. Ether's price saw its first significant drop at the end of 2016, following "The DAO" incident, and Ethereum was in a recovery phase. In March 2017, the Enterprise Ethereum Alliance was formed, and the price of Ether rose to $40, then experienced another decline. With the surge in Ether's price, block rewards and transaction fees ensured Ethereum's stable operation, leading to a low point in miner balances. Over the past year and a half, miner balances seem to have been increasing. In the last 18 months, the amount of Ether held by miners has increased from 600,000 to 1.6 million. These individuals are true hodlers. However, to study the behavior of non-holders, we need to understand the situation in exchanges.

Amount of ETH held by various exchanges

Value of ETH held by various exchanges in USD

A significant factor contributing to Ethereum's scale and success is its thriving exchange ecosystem. Unlike the current new coin listings, Ethereum had low listing requirements on major exchanges like Kraken and Poloniex. If Ethereum had not reduced the barrier for asset issuance and built a community around new assets, we would not see the emergence of nearly 1,600 digital assets in the market today. It's hard to imagine what the token ecosystem would look like without Ethereum. When I combined these charts, I noticed a critical data point: the circulating supply of Ether on exchanges is lower than that of Bitcoin. According to Glassnode data, Bitcoin's circulating supply is 18 million, with exchanges holding about 2.1 million (approximately 11.6%). In contrast, Ethereum's circulating supply is 109 million, with exchanges holding about 9 million (approximately 8.25%), slightly lower than Bitcoin. This is likely because Ethereum transactions are relatively faster and cost less. Due to Ethereum's faster transaction speeds, people find it easier to transfer assets and profit during price fluctuations.

Note: The actual amount of Ether held by exchanges may be slightly higher by a few percentage points as my data source does not include Coinbase's balances.

After researching this data, I have drawn two additional conclusions:

1. Since 2016, the amount of Ether held by exchanges has decreased by 50%. This may indicate a decreasing number of people treating Ether as a speculative asset.
2. Currently, the total value of Ether held by exchanges is $1.5 billion. In comparison, at the time of writing, the value locked in decentralized finance smart contracts is half that of exchanges, at $875 million. When the value of Ether locked in decentralized finance exceeds that held by exchanges, we will see a significant shift.

Next is the final indicator reflecting the current state of the Ethereum ecosystem—ICO balances. However, this indicator raises some concerns.

Largest ICO wallets as of October 2019

During the ICO frenzy, major projects raised approximately $8 to $16 billion (figures vary based on different sources). Currently, the value of Ether held by major ICO projects is around $500 million. Over the past two years, these projects have been burning through their funds and seem to have little left. Take EOS, for example, despite raising $4 billion and spending $30 million to acquire http://voice.com, it has failed to build its own social media platform on the blockchain (possibly planning to build it on BitShares, hehe 🤷). If the network does not generate revenue or the platform fails to create income, the projects can only seek risk capital or face closure. By the end of 2019, we have already seen many projects closing down. In 2020, if these altcoins disappear, ICO projects go bust, and assets shrink significantly, these projects will face major challenges. This is primarily due to:

1. Over the past two years, these projects have been burning through funds without much progress.
2. The performance indicators of these ICO projects will deteriorate over time, and without sudden growth like MakerDAO, later-stage funding prospects do not look promising.

The bleak outlook for projects, lack of investment, and low market adaptability of products (such as Bancor and Uniswap), combined with dwindling funds, pose a significant challenge for the industry and community to reflect on. While this may not necessarily be a bad thing, a cash crunch often forces entrepreneurs into survival mode, focusing on key performance indicators rather than burning through capital. This may also prompt entrepreneurs to conduct experiments they have never tried before and explore new business models. In the coming years, we are likely to see a wave of project closures.

Amount of ETH locked in various DeFi projects

Decentralized finance has grown 30 times since 2017, largely due to Maker. A key reason for this growth is the need for stable assets as a hedge against token price volatility. Since the emergence of Tether, stablecoins have become popular. Additionally, projects like MakerDAO allow Ether whales to leverage their holdings through Maker when Ether's price hit a new low (i.e., $80). Since 2018, the decentralized finance sector has seen substantial growth. Interestingly, the locked Ether in decentralized finance projects exhibits a clear power-law distribution. The locked Ether in the decentralized finance sector is mainly used to generate SAI and DAI, with Compound ranking third. Trading tools like Uniswap and DyDx have also attracted some Ether, but not as much. The market seems to have decided on its preferred currency issuance protocol. Projects like Synthetix are experimenting with new models, utilizing Ethereum and the SNX token to issue assets. Currently, Synthetix's development is promising. From the chart, it is evident that this industry requires new business models. Relying solely on locked Ether and transaction fees from projects may not be enough to sustain a profitable business. Hopefully, we will see more new projects in 2020.

Github activity of several mainstream projects (top line is Ethereum, followed by purple for Bitcoin, then EOS, and the bottom blue line is Maker)— Source: Santiment

Ultimately, what sets Ethereum apart is its user base. The more users Ethereum attracts, the richer its network becomes. When people think of Ethereum and its community, they often recall controversial events (such as ICOs, "The DAO" fork, etc.) that seem unacceptable to them. However, they often overlook the fact that these events occurred because the Ethereum network can attract a diverse range of people, most of whom are developers. Compared to other networks (and their Github activity), Ethereum has consistently been ahead, even during the long bear market. It is this culture of innovation on Ethereum that has led to the emergence of numerous innovative projects in decentralized finance, non-fungible tokens, and token issuance. As long as this momentum is maintained, Ether will have a greater impact on the market. Whether this impact is worth $10, $100, or $1,000, I do not know. The efficient market hypothesis and time will tell us the answer.

Despite Ethereum's journey filled with challenges, I believe the network is continuously progressing. Some may think Ethereum is overrated, but no chain has brought forth innovations like Ethereum, encompassing ENS, decentralized finance, non-fungible tokens, and ICOs. Before Ethereum becomes mainstream, there is still a long way to go for Ethereum and its community. Yet, as I engage with the Ethereum community and interact with Ethereum's founders, I increasingly feel a resemblance to Ubuntu/Linux in 2007-200