Is the influence of miners on the Bitcoin network weakening? On-chain data also reveals these trends

Original article: coinmetrics.io

Written by Karim Helmy and the team at cryptocurrency data provider Coin Metrics
Compiled by: Perry Wang
Coin Metrics authorized Lianwen to translate and publish the Chinese version of this article.

Key Points

• This report uses a new approach to quantify miners’ holdings and activity by looking at addresses in coinbase transactions. This approach improves on previous studies that tracked miners’ spending, which inadvertently measured the activity of mining pool operators rather than the behavior of miners.

• In the year before the halving of Bitcoin production, as the price of the currency recovered from the lows in recent years, the number of Bitcoins accumulated by miners increased by another 318,000 BTC.

• As miners’ supply dwindles over time and the net flow of funds to their addresses stabilizes, miners appear to have less influence on the Bitcoin network.

• The upcoming Coin Metrics Network Data Pro 4.8 release will provide miner fund flow and supply metrics .

Miners and Markets

In addition to their role in securing the network, miners have a profound impact on Bitcoin’s market dynamics. Because they can receive newly issued Bitcoins rather than purchase them, miners are naturally net sellers of the asset . This effect is further enhanced by the fact that miners’ operating expenses (primarily electricity and rent) are primarily denominated in fiat currency, while their income is denominated in Bitcoin.

The new functionality leverages previously unavailable data about accounts interacting with these addresses to examine miners’ activity, assessing their drivers and spending impact.

On-chain data suggests that miners’ influence on the network is gradually decreasing , but they are still key players in the Bitcoin ecosystem with a lot of money. To help our readers understand these influencing factors, Coin Metrics provides extensive data related to miners in the upcoming Network Data Pro 4.8 release. Using this data, the feature finds that the Bitcoin supply held by miners generally decreases over time , while the flow of funds in and out of miners and mining pools has been suppressed by the network’s successive halvings.

Bitcoin supply in a nutshell

To calculate the miner money flow, we first aggregate all addresses that received payments from coinbase transactions (Chain News Note: coinbase transactions are the first transaction in a block, a unique type of Bitcoin transaction that can be created by miners, which miners use to collect block rewards for their work, and any other transaction fees collected by miners are also sent in this transaction) and mark them as first transfer (0-hop) addresses. Then, all addresses in this group and addresses that have received payments from an address in this group are marked as second transfer (1-hop) addresses.

Since the usual architecture of mining pool wallets is that the mining pool is the first to receive the rewards in the block and then distributes them to the miners, the 0-hop address usually represents the mining pool , while the 1-hop address usually represents the miner . It can be seen from this that existing systems that attempt to infer miner behavior from 0-hop address spending habits are theoretically incomplete and therefore cannot accurately measure miners' intentions. Instead, they measure the activities of mining pool operators .

Of course, it is acknowledged that labeling miners and mining pools based solely on the distance in the coinbase transaction is an imperfect technique. This methodology works well when applied to assessing the early Bitcoin network, where individual mining and alternative mining pool models were more popular. Since the first mining pool , Slush Pool , mined its first block in December 2010, measurements before that date in particular are only for reference. In addition, miner addresses that have not yet received funds from a 0-hop address are not labeled. Nonetheless, this heuristic represents a significant improvement over the current state of the art and should accurately capture broad trends.

Miners, especially those active in the early days of the network, control a large amount of Bitcoin . Throughout the history of the Bitcoin network, the total number of Bitcoins owned by 0-hop and 1-hop addresses has generally been on a downward trend. In the second half of 2019 and the first half of 2020 (i.e., the year before the halving), this trend reversed significantly, with miners accumulating 383,000 BTC from bottom to peak. This effect was mainly limited to 1-hop addresses, while the 0-hop supply remained roughly the same. Therefore, most of this accumulation would not be detected by previous estimation techniques.

As can be seen in the graph, there have been several jumps in the supply held by miners. These spikes are usually caused by the first block mined by an address with a large balance, or the first interaction with a previously marked 0-hop address. The most prominent of these spikes occurred on August 16, 2012, when a whale holding over 500,000 BTC received part of the coinbase reward for block 194,256. New entrants were also one of the reasons for driving the supply controlled by miners higher before this year's halving.

The gradual reduction in supply owned by miners and mining pools is even more significant from the perspective of total supply due to inflation. This decline is consistent with the overall trend of increasing dispersion of Bitcoin supply. It is also consistent with the trend of increasing popularity of the mining pool model, which means that non-mining addresses are less and less likely to be overly marked as 1-hop addresses.

Despite this trend, miners and mining pools still control a large portion of the total Bitcoin supply even today.

Analyzing mining pools and transfers

The flow of funds between these groups is another strong on-chain signal. Because mining pools are often direct recipients of coinbase rewards, 0-hop address fund flow is a useful indicator of mining pool activity. In particular, there have been several spikes in activity, represented by the activity of the whales mentioned above. In addition, since the early days of the Bitcoin network, the inflow and outflow of 0-hop addresses in BTC has been on a downward trend .

Miner income or block reward income accounts for the bulk of the inflow to 0-hop addresses. Although miner income varies in the short term due to fluctuations in fees and the number of blocks mined, it is relatively stable across periods.

While inflows and outflows are highly correlated, outflows are more volatile because they are not backed by protocol rules and miners can choose when to withdraw funds from pool wallets. The impact of the 2016 and 2020 halvings is evident in the decline in both flows. Since the 2020 halving, the value of inflows has generally exceeded the value of outflows, a reversal of historical norms.

Analyzing Miner Funds

While 0-hop address fund flows are useful for tracking payments to pool operators, they do not represent transfers made by miners themselves in today’s standard wallet architecture. In most mining pools, block rewards are received by an address controlled by the pool operator , who holds the funds in escrow until the miner is paid at predetermined intervals or chooses to withdraw the tokens.

In the pooled mining model, the fund flows from the 0-hop address of the coinbase transaction are closer to the miner's expenditure . This report is one of the earliest studies to try to analyze the fund flows of 1-hop addresses. Due to the larger number of research addresses, lack of underlying support, and fast fund flow, these fund flows are much larger and more volatile than the corresponding fund flows of 0-hop addresses.

To analyze the flow of funds in the early days of the Bitcoin network (when mining pools had not yet become the mainstream mining model), 0-hop address fund flow may be a more appropriate tool. Even today, 1-hop address fund flow is only an approximation of miner activity because the wallet structure of mining pools varies and exchange addresses may be erroneously included in these fund flows. Overall, however, this research model represents a more comprehensive view of miner spending under today's network conditions.

Like 0-hop address capital flow, 1-hop address inflow and outflow are also closely correlated. Since block rewards only represent a small part of 1-hop address inflow, both inflow and outflow are very volatile, so the impact of Bitcoin halving on this type of capital flow is not obvious .

Over the past year or so, the inflow and outflow of miner addresses have shown a slight increase, indicating an increase in activity. Because the net capital flow has remained roughly stable and volatility has decreased overall, the increase in activity will not be reflected in an increased impact on the network.

The close relationship between inflows and outflows suggests that miners, in general, tend to move their owned Bitcoin out of their addresses immediately. Given that derivatives markets and fiat lenders often require custody of tokens, these flows do not preclude miners from using financial instruments to hedge their exposure to Bitcoin prices. However, survey data from the Cambridge University's Third Global Cryptoasset Benchmark Study suggests that the adoption of these financial instruments is low , with miners relying primarily on holding and selling Bitcoin to keep their risk levels to a desired level. Such high transaction volumes seem to indicate that miners are active market participants who habitually sell most of their newly received Bitcoin.

Impact of US dollar pricing

Since miners’ expenses, profits, and losses are all denominated in USD , it is useful to measure miners’ fund flows in USD terms. Since 0-hop address fund flows are mainly composed of block rewards, their USD value curve is very similar to the miners’ income curve .

Miner flows in USD are similar to mining pool flows (which are much larger) due to their similarity to Bitcoin prices. However, unlike mining pool flows, miner flows have been on an upward trend , even briefly exceeding the level of Bitcoin's all-time high in 2017 in late 2019. This suggests that miner activity across the network is more extensive in USD terms.

Comprehensive considerations

Taken alone, the inflow and outflow of funds to an address are useful for measuring the amount of economic activity that miners are involved in.

For most of the Bitcoin network’s history, net fund flows from 0-hop addresses have been slightly negative, with these addresses typically spending more than they receive. Although net fund flows were volatile in the early days of the network, they have become less volatile over time, likely due to the halving of Bitcoin production .

The volatility of net fund flows to 0-hop addresses has continued to decrease in recent years. In the past few months, the historically negative value of net fund flows has reversed, with inflows slightly greater than outflows since the most recent halving.

The net fund flows of 1-hop addresses are more volatile than the net fund flows of 0-hop addresses. Like the net fund flows of 0-hop addresses, the net fund flows of 1-hop addresses are generally negative. These fund flows have also experienced a gradual decrease in volatility, indicating that miners have a gradually decreasing impact on liquidity .

Another tool we use to analyze fund flows is the Miner Inventory Level Change or MRI, which is the ratio of miner outflows to miner inflows. The 0-hop address MRI can be used to measure whether miners are depositing funds into mining pool wallets (less than 100% ratio) or withdrawing funds (more than 100% ratio).

Since 0-hop address fund flow is closely related to miner income, MRI has remained close to 100% in most of the Bitcoin network's history. The volatility of this indicator has gradually decreased as the volatility of 0-hop address outflows has weakened.

Compared to 0-hop address MRI, 1-hop address MRI can provide a more accurate picture of the relationship between miners’ spending habits and income. Throughout the history of the Bitcoin network, 1-hop address fund flows have remained relatively stable, and after two consecutive halvings that reduced issuance, 1-hop address MRI has risen by leaps and bounds. Since 1-hop address spending is about an order of magnitude higher than miners’ income, 1-hop address MRI has reached thousands of percentage points at many times throughout the network’s history.

Next steps and actions

On-chain indicators such as miner holdings and net transfers show that miners' influence on the Bitcoin network is gradually decreasing. Despite this, they still contribute a lot of activity and control a large part of the total Bitcoin supply. Some indicators, such as total fund flows , also suggest that miner activity is increasing in both US dollars and Bitcoin.

As the only direct recipients of Bitcoin issuance, miners and mining pools have an impact on the network that is not easily quantifiable, and the metrics outlined in this article only scratch the surface of miners' behavior . In the future, we hope to analyze the flow of Bitcoin funds from miners to exchanges, which can more directly measure its impact on the market. We also hope to evaluate the active supply held by miners' addresses, which will help filter out Bitcoin lost in the early days of the Bitcoin network, and take into account the different wallet structures of various mining pools, ultimately providing a more detailed assessment of miners' behavior.

Thanks to Celia Wan for suggestions and editing.