TBW - Bitcoin Mining: A Profitable Industry for the Powerful Few
Since the very first bitcoin was mined in the Genesis block by Satoshi Nakamoto on 3 January 2009, an industry has grown up around the idea that computing power can be converted into an extremely rare digital reserve. In the early days, mining was an accessible activity that could be carried out on a personal computer equipped with a simple graphics card. The initial reward for each block mined was 50 bitcoins. Every four years, this reward is halved, in line with the halving mechanism written into Bitcoin's code. The first reduction took place in 2012. Today, the reward is 3.125 bitcoins per validated block. Despite this gradual drop in earnings, the mining sector is as active as ever.
At the start, the ratio between the reward obtained and power consumption was largely favourable. During the first halving cycle, bitcoin peaked at $26.90. At that price, a mined block was worth around 1,345 dollars, for a limited production cost and a computer running on Bitcoin Core. The difficulty level of the network, which reflects the cumulative effort of all participants seeking to validate a block, was then virtually zero: 8e-12 EH/s, 14 orders of magnitude less than today's hashrate.
As the price of bitcoin rose and its adoption grew, new miners joined the network, boosting its computing power, its security (a malicious actor needs more resources to carry out a 51% attack), but also its energy consumption. Competition between miners has intensified, pushing them to regularly upgrade their hardware. Traditional chips have rapidly become obsolete, giving way to ASICs (Application-Specific Integrated Circuits), machines dedicated solely to mining. More powerful, they also consume more electricity. Rising prices have boosted bitcoin's appeal as a monetary asset, attracting new players willing to pool their power in pools to maximise their chances of success. Some companies have even raised funds on the stock market to finance industrial mining farms.
This report provides an overview of the Bitcoin mining industry, its evolution, its characteristics and the risks associated with this type of activity.
The fundamentals of Bitcoin mining
Bitcoin mining is based on a consensus mechanism called proof-of-work, in which miners mobilise computing power - measured in hashes per second - to try to guess a value called nonce. The latter, a 32-bit number used only once, must result in a hash lower than a target value set by the protocol. To achieve this, the miners add this nonce to the various elements of the block (transactions, timestamp, hash of the previous block, Merkle root, software version) and submit the whole to the SHA-256 hash algorithm. The result, a hexadecimal code, is then compared with the target: if it is less than or equal to the target, the block is validated and added to the blockchain. After 6 to 10 additional blocks, the new block is considered virtually immutable.
The announcement is the only parameter of the block that the miner can modify. Thus, the more computing power he has (the more hashes he can make per second), the more he multiplies his chances of finding a valid result and therefore entering a block in the chain - and collecting the associated reward.
On average, a new block is produced every 10 minutes. To maintain this rate, the protocol automatically adjusts the difficulty of mining every fortnight, or every 2016 blocks. If the number of miners increases, or if more powerful hardware is used (such as new ASICs), blocks may be found more quickly than expected. In this case, the difficulty is raised to slow the pace. Conversely, if miners leave the network and computing power decreases, the difficulty is lowered, making mining easier for the remaining participants.
A global but concentrated industry
Bitcoin mining is a globally distributed activity. The network has active nodes on every continent, with a significant concentration in the United States, Europe, China and Hong Kong. There are currently 16,525 nodes connected to the network. Of these, 62.5% are active solely on the Darknet, 17% on the light network, and the rest alternate between the two types of connection.
This distribution of nodes reflects a geographical decentralisation of the network. But a closer look at the organisation of the mining pools reveals a very different picture. The seven main pools alone accounted for 86.15% of total bitcoin production over the past week. Together, these pools had a combined computing power of 779.08 EH/s, while the overall hashrate was 806 EH/s. In other words, 96.6% of the mining power is concentrated in the hands of these seven entities.
The Foundry USA pool dominates activity, with 281 blocks mined, corresponding to a reward of 878.125 BTC, or around $93.3 million. It is followed by AntPool, which generated 593.75 BTC, or around $63.1 million over the period. These two pools concentrate a significant share of the market, reflecting the industrial structuring of mining.
It is important to note that mining pools are not limited to miners from their country of origin - anyone can connect to them. The precise geographical distribution of miners is therefore difficult to establish, as they can use VPNs to mask their location. Moreover, countries do not yet have specific customs codes to track the importation of ASICs into their territory.
Despite China's official ban on mining since May 2021, several sources report the existence of clandestine operations. The presence of AntPool among the main mining pools and Bitmain as a leader in mining equipment manufacturing (both of which are part of the same industrial group) confirms that China remains one of the major players in the sector, contradicting Beijing's official position.
At first glance, the performance of the main US and Chinese pools seems comparable, with an efficiency of around 3.619 BTC per exahash. But some of the smaller players, spread across the rest of the world, are performing much better. With just 3% of global computing power, these miners generated 14% of total bitcoin production over the same period. Their efficiency reached 16.665 BTC per EH/s, almost five times that of the market leaders. A gap that raises questions about the cost structure, optimisation of resources and operating conditions of these more discreet players.
The economic balances of Bitcoin mining
Bitcoin miners are rewarded in bitcoins each time they successfully validate a block. This reward is made up of two elements: a fixed subsidy, defined by the protocol (currently 3.125 BTC per block), plus the transaction fees included in the block concerned.
The majority of miners' income comes from this subsidy. In the first quarter of 2025, it accounted for between 98% and 99% of their revenues. On May 25, for example, the miners' total revenues reached $49.32 million. Of this, $48.62 million (or 98.5%) came from the subsidy, compared with just 1.5% from transaction fees. Despite a gradual decline in the reward in bitcoin - due to the halving mechanism - and declining unit returns, miners' overall revenues have shown an upward trend for the past three years, although they remain below the peak reached in 2021.
On a global scale, the average cost to produce one bitcoin is estimated at $90,304. At the time of writing, the bitcoin price is around $110,041, a margin of 0.82. If this ratio exceeds 1, it means that the average production cost is higher than the value of the bitcoin mined, and that miners are operating at a loss. Since November 2024, most miners have remained in the black, with the exception of April when the ratio exceeded 1.10. On average, a miner can currently generate a margin of $20,000 per bitcoin produced, provided they sell it immediately after mining.
However, this profitability remains fragile. Strong competition between players is pushing up technical requirements, forcing miners to deploy ever more computing power to hope to add a block to the chain. Energy inflation in recent years has also weighed on costs. While equipment has become more efficient - fewer joules per terahash are needed - the sector remains highly capital-intensive. To remain competitive, miners have to invest regularly in new machinery or locate cheaper energy sources than their competitors.
Energy consumption and environmental impact
The energy consumption and CO₂ emissions associated with Bitcoin mining are regularly the subject of criticism. The industry currently consumes around 176 TWh (terawatt hours) a year, equivalent to the electricity consumption of a country like Poland. It generates around 98.1 million tonnes of CO₂, consumes 2,772 gigalitres of water and produces 20.34 kilotonnes of electronic waste every year.
When it comes to energy sources, the mix remains mixed: 42.6% of the electricity consumed comes from renewable sources, with hydroelectricity predominating. Non-renewable energies account for 47.6% of consumption, dominated by coal. Nuclear power, meanwhile, accounts for 9.8%. When renewables and nuclear are combined, the share of electricity from low-carbon sources rises to 52.4%.
Despite the overall rise in energy consumption, the mining industry is moving towards greater efficiency. The most efficient machines today consume between 12 and 42 joules per terahash (J/TH). The average efficiency is 21.81 J/TH. This technological gain has enabled total network power (hashrate) to grow by 455% since January 2021, while electricity consumption has only increased by 111% over the same period.
On a global scale, the 176 TWh used annually by miners represents a modest fraction of total energy consumption. The activity absorbs 2.18% of global industrial consumption, 1.88% of household consumption, 3.41% of the commercial sector, and between 44% and 73% of data centre consumption, according to available estimates. In other words, although visible and energy-intensive, Bitcoin mining remains far from a major item in the global energy footprint even if it is substantial.
Technological trends and infrastructure evolution
As of 24 May 2025, the Bitcoin mining industry has seen notable advances in terms of infrastructure, whether in terms of hardware, cooling systems or large-scale mining farms. The increasing difficulty of the network, the search for better energy performance and the need to adapt to sustainability issues have all contributed to accelerating this transformation. The transition from an amateur model to an industrial activity is now clearly visible through the increasing sophistication of installations.
Mining machines, now largely dominated by ASICs (application-specific integrated circuits), have seen rapid gains in efficiency. Manufacturers such as MicroBT are offering new-generation models - such as the Whatsminer M56 and M53 - designed for cooling systems adapted to heat-dense environments, such as immersion or hydraulic cooling. These machines aim to lower energy consumption per terahash, improving profitability and reducing environmental impact. At the same time, cooling technologies are diversifying, with the emergence of single- and dual-phase immersion solutions, complementing traditional air- and water-based approaches.
Mining farms, meanwhile, are growing larger and more complex, incorporating state-of-the-art infrastructure to optimise performance, location and energy consumption. Some companies, like Core Scientific, are now exploring related activities such as artificial intelligence and high-performance computing (HPC), using shared facilities. There is also a growing trend to locate these farms in cold zones or close to renewable energy sources, with a view to reducing operating costs and the environmental footprint.
This move towards a more efficient infrastructure is based on a technological convergence: increasingly high-performance ASICs require advanced cooling systems, supporting scalable and sometimes multi-use mining operations. Yet a number of challenges remain: regulatory requirements, hardware compatibility and sustainability constraints will continue to shape the future of this industry.
Performance of listed companies
Among listed mining companies, Marathon Digital Holdings was the most productive in the first quarter of 2025, with a total of 2,285 bitcoins mined. This result can be explained in particular by the fact that MARA operates its own mining pool and has the most computing power among the public players, with 57.3 EH/s deployed. In second place, CleanSpark produced 1,956 BTC with a hashrate of 42.4 EH/s. RIOT rounded out the podium, with 1,530 BTC mined and a power of 33.7 EH/s.
In terms of operational efficiency - measured in BTC produced per EH/s - it was Iris Energy (IREN) that stood out, with a ratio of 51.50 BTC per EH/s. Conversely, Hut 8 had the lowest efficiency, with 17.96 BTC per EH/s. IREN was also the most invested group over the period, with $442 million allocated to acquiring and upgrading its fleet of machines, which enabled it to post the lowest marginal cost per bitcoin produced in the first quarter.
CleanSpark ranked second on the operational efficiency criterion, with 46.13 BTC per EH/s. Combining this performance with modest capital expenditure over the period, the company consolidates its position as one of the most efficient listed companies in the sector.
Bitcoin mining as a strategic lever for states
Bitcoin mining represents a strategic opportunity for states seeking to diversify their economies, particularly those with access to cheap or surplus energy. Countries such as Paraguay and Argentina are already exploiting their hydroelectric resources or natural gas surpluses to mine Bitcoin, turning underused energy into economic value. This leverage can be used to generate alternative income and strengthen financial stability, while reducing energy waste.
On the US side, this dynamic is part of a broader rationale of economic resilience, embodied in the Strategic Bitcoin Reserve policy, which envisages the integration of bitcoin as a strategic asset. The issue ceiling of 21 million bitcoins makes it an attractive tool for building up national reserves in the face of inflation or currency erosion. Mining allows bitcoins to be acquired directly, without going through the markets, thereby strengthening the financial autonomy of the states concerned.
Aside from the economic aspect, the development of mining can also encourage a rise in technological skills, particularly in the fields of blockchain, energy management or high-performance computing (HPC). This can help create jobs and stimulate innovation, particularly in countries where the digital infrastructure is still in a growth phase. With 42.7% of the world's hashrate concentrated in its main pools, the US sees mining as a pillar of its position in the global digital economy. Encouraging other countries to adopt this model would contribute to a better geographical distribution of the Bitcoin network and a more rational use of renewable or isolated energies. This is already the case in Norway and Ethiopia, which rely on hydroelectricity, and Argentina, which uses gas flaring for mining. Criticism persists, however, particularly in the United States, where some studies point to the environmental impact of the activity.
On a geopolitical scale, wider participation in mining can be seen as a factor in financial decentralisation and resilience, while maintaining a form of influence in the new digital economic order. Several political figures have expressed their desire to preserve American leadership in this field.
An often underestimated use case for mining concerns the stabilisation of electricity grids. A mining farm can act as a balancing mechanism, absorbing surplus production to avoid losses or overloads on the grid. In some contexts, mining Bitcoin is more profitable than selling surplus electricity to neighbouring countries. Conversely, in the event of tension on the grid or increased difficulty, the installations can be paused. Some infrastructures are even designed to produce or store electricity, or to be converted back into data centres, particularly for applications linked to artificial intelligence.
Regulation: a rapidly changing framework
As of 26 May 2025, national political developments and international environmental pressures are gradually reshaping the regulatory framework around Bitcoin mining. In the United States, the growing institutional adoption of bitcoin and the strategic bitcoin reserve initiative are evidence of the normalisation of this activity within economic policy. For all that, certain obstacles persist, particularly in connection with the compliance requirements introduced by legislation such as the Infrastructure Investment and Jobs Act (IIJA). An interdepartmental working group, led by a special adviser on AI and cryptoactives, illustrates the government's desire to structure a clearer and coordinated framework.
Internationally, mining regulation is evolving in a fragmented way. Some countries, such as Belarus, are exploring mining as a strategic tool for extracting value from energy surpluses, while others, such as Kazakhstan, are stepping up supervision in the face of pressure on power grids. Australia, meanwhile, is moving towards wider recognition of the sector by adapting its infrastructure and regulation for digital assets. Despite this diversity of approaches, the environmental impacts of mining remain a major global concern.
By 2030, the share of renewable energy used by the industry is expected to increase significantly, driven by a combination of technological advances and political incentives. This change opens up opportunities, but also risks for operators. International miners face a contrasting landscape, between favourable countries and more restrictive environments. In the United States, regulatory uncertainty coexists with significant growth potential, provided they know how to adapt to new requirements, particularly in terms of energy consumption.
As the mining sector aligns itself with broader economic and environmental objectives, players who favour efficient technologies and a sustainable approach are best placed to come out on top.
Risks associated with Bitcoin mining
Bitcoin mining involves a number of risks, particularly for players seeking to maintain a profitable business model in an ever-changing environment. The main risk is financial: operating costs are high, and returns are never guaranteed, particularly for older or less efficient machines. In periods of high energy demand, electricity prices can rise rapidly, squeezing miners' profit margins even further.
The market risk is also structural: the price of bitcoin can remain below the average cost of production for several months, as was the case between April and October 2024. Under these conditions, the business rapidly turns into a loss-making operation. Added to this is regulatory uncertainty. Pressure from the authorities can change abruptly, culminating in total bans like the one imposed by China in 2021. Environmental concerns - CO₂ emissions, production of electronic waste, overheating of facilities or overloading of power grids - may also prompt some regulators to restrict or ban activity.
Finally, the security risk should not be overlooked. As mined bitcoins are often held by the miners themselves or by institutional custody structures, the threat of theft or intrusion remains very real. This factor, combined with other structural uncertainties, serves as a reminder that mining is an exposed business, requiring rigorous risk management and a capacity for continuous adaptation.
The Big Whale's view
One thing is clear: Bitcoin mining remains a profitable business, but only for those with the capital to upgrade their machines, adopt high-performance cooling systems and make their infrastructures ever more efficient. The margin for profitability has narrowed considerably, and at current price levels it is difficult to maintain widespread profitability over the long term. Market volatility regularly pushes the BTC price below the break-even point, pushing marginal costs above 1, resulting in losses. Only the most efficient operators manage to maintain stable activity, hence the strategic importance of having the best equipment.
As competition intensifies and hashrate continues to grow across the network, staying in the race requires significant investment. This financial barrier is increasingly pushing the business towards an institutional model, where access to capital enables a technological advantage to be maintained. Faced with these challenges, a number of listed players are redirecting part of their capacity towards high-performance computing (HPC), by leveraging their data centres and computing power to generate additional revenue.
On a national scale, mining can also be of strategic interest. It offers a way of accumulating bitcoins for those with energy surpluses, while also helping to stabilise power grids. In some countries, this activity could even encourage the acceleration of the transition to renewable energy, by directly converting clean energy into a monetary reserve.
The capital-intensive nature of the sector and the increasing complexity of mining highlight the importance of hardware development. The objective is clear: to perform more calculations with less energy, and thus reduce the cost per exahash. In a context where the difficulty of mining has increased by 455%, where the subsidy per block decreases with each halving, where bitcoin is growing by an average of 42% a year, but where electricity costs are also rising, players have two options: invest to increase their computing power and maximise their chances of reaping rewards, or keep old machines, try to make them more frugal, and hope that one day they will mine a block.
