What a Bitcoin Blackout Reveals About Mining Resilience

The pioneering digital asset, Bitcoin (BTC ), is decentralized at the protocol layer. However, mining remains physically concentrated around low-cost power.

The Xinjiang blackout demonstrated just how a local energy disruption can temporarily become a network-wide stress event, raising settlement costs, delaying transactions, and weakening market quality.

While the Bitcoin network continued to operate as intended, a prolonged power blackout caused substantial disruption to the digital asset’s global computing power before recovering without centralized intervention. Such events raise an important question: just how resilient is Bitcoin’s mining infrastructure when large concentrations of computational power suddenly go offline?

As institutional investment in the cryptocurrency continues to expand and mining becomes more industrialized, the answer has significant implications for network security, market stability, and the strategic direction of major mining companies.

The Hidden Weakness Behind Bitcoin’s Decentralization

Over the past decade, cryptocurrencies have seen a considerable increase in popularity, and among all crypto assets, Bitcoin continues to lead in adoption.

The most prominent crypto asset gave rise to the entire industry. Bitcoin was the first to operate without any central authority like a bank or clearinghouse. Instead, it relies on a network of specialized computers, known as Bitcoin miners, spread out all over the world.

These miners validate transactions, group them into blocks, and secure the blockchain through Bitcoin’s Proof-of-Work (PoW) consensus mechanism, which addresses the double-spending problem of decentralized crypto assets without trusted authorities.

To get a chance to verify transactions, miners compete by solving computationally intensive mathematical puzzles, and the first one to find a solution gets to add the next block of transactions to the chain. In exchange, miners earn newly issued BTC along with transaction fees. This whole process is very energy-intensive, so miner profitability comes down to the efficiency of their hardware and the cost of the electricity running it.

It’s only natural, then, that miners gravitate toward regions with cheap electricity, whether that’s stranded hydroelectric capacity, surplus wind generation, or heavily subsidized coal power. Over time, this creates geographic clustering that undermines the network’s decentralization.

As the crypto market continues to evolve, become institutionalized, and gain regulatory clarity, concerns regarding the limits of Bitcoin’s decentralization and the ecological impact of mining are increasing.

Bitcoin is built to have no single point of failure, but that requires geographically distributed and economically competitive miners, making it difficult for any single entity, region, or external event to compromise the network. But in order to be profitable, miners need cheap power, which means much of the mining capacity sits behind the same regional power grid.

This concentration of mining capacity in a handful of locations or among a small number of operators produces several large points of failure. For instance, localized disruptions caused by power outages, natural disasters, geopolitical tensions, or regulatory actions can negatively affect the network’s ability to process transactions efficiently and its security.

This is where mining resilience comes in.

Interestingly, Bitcoin has a built-in shock absorber in the form of mining difficulty, which automatically adjusts roughly every two weeks to keep block production at about one every ten minutes.

As explained in Bitcoin’s whitepaper, “To compensate for increasing hardware speed and varying interest in running nodes over time, the proof-of-work difficulty is determined by a moving average targeting an average number of blocks per hour. If they’re generated too fast, the difficulty increases,” and vice versa.

What this means is that if a large chunk of hashrate suddenly disappears due to any disruption, the network slows down, but only temporarily, as the system self-corrects within a couple of weeks.

The protocol is designed to withstand failures, but the real-world infrastructure remains vulnerable to electricity shortages, grid failures, and regional disruptions. So, what exactly happens in the meantime, during the period before the network adjusts, when a chunk of mining capacity concentrated in a specific location goes dark?

To answer that, we need to examine such disruptions, and a large-scale blackout provides the perfect opportunity to observe how the network performs under genuine operational stress. A recent study has done exactly that, offering valuable insights into both the strengths and limitations of Bitcoin’s decentralized architecture.

To analyze the risks associated with the geographical centralization of mining, the researchers used an exogenous shock to the electricity supply in a relatively small region with heavy Bitcoin mining activity.

The study also compared the effects of the blackout on Bitcoin to a cryptocurrency using a more energy-efficient consensus mechanism.

What the Regional Blackout Revealed About Bitcoin Mining

“Many cryptocurrencies are designed to be decentralized, but in practice market forces may drive them towards lower degrees of decentralization,” noted the study titled “Bitcoin blackout: Proof-of-work and the risks of mining centralization¹,” which examined the impact of a multi-day blackout on the network and provides an empirical analysis of how concentrated mining infrastructure affects PoW resilience.

Instead of relying on simulations, researchers used an actual disruption for their analysis. The disruption took place in Xinjiang, China, which temporarily knocked out a significant portion of the Bitcoin network’s computational capacity.

Authors Stefan Scharnowski from the University of Mannheim, Germany, and Yanghua Shi from the International University of Japan, Japan, found that the blackout reduced Bitcoin’s global hashrate by roughly 24%.

As economist Alex de Vries noted in an interview at the time, besides halting computing power, “we also learned that the area in Xinjiang where all that mining happens is much smaller than previously believed.”

The study also pointed out that not only did Xinjiang account for one-third of worldwide mining activity, but even within the region, all mining operations are located in close geographical proximity.

It further mentioned a more recent event, Kazakhstan’s 2022 country-wide internet shutdown, which resulted in a 15% drop in Bitcoin’s hashrate.

These events demonstrated that a considerable share of mining power remained geographically concentrated enough for a regional infrastructure failure to have measurable network-wide effects.

To isolate the effects of this shock, the researchers compared Bitcoin’s behavior to that of altcoin ADA. Cardano’s (ADA ) native token uses a proof-of-stake (PoS) consensus mechanism, which is far less energy-intensive than PoW and therefore does not depend on cheap regional electricity, but is subject to other forms of concentration like software dependencies or validator stake concentration that “may give rise to similar systemic vulnerabilities.”

The results showed that during the blackout, Bitcoin’s implied hashrate fell from around 170 exahashes per second (EH/s) to about 130 EH/s. This drop was consistent with the shutdown of one large or several mid-sized coal power plants in the region.

The decline in computing power led to a slowdown in Bitcoin block production. With fewer miners to find valid blocks, transaction settlement times temporarily increased. On average, Bitcoin transactions are confirmed within 10 minutes, but this jumped to about 16.4 minutes during the blackout.

But one confirmation isn’t enough. Most platforms require three to six block confirmations. So, traders had to wait about 38 minutes for their standard six-block confirmation, roughly 64% longer than normal, to be confident that a transaction was irreversibly settled. This was “a statistically and economically significant increase in settlement latency,” the study said.

This congestion showed up in the mempool, a waiting room for pending and unconfirmed transactions, which surged as high as 218,000 by the end of the event, a threefold increase from the stable baseline of around 60,000. The study pointed out:

“The average of about 150,000 during the blackout period is statistically significantly larger than before. Afterwards, the number of transactions waiting in the mempool decreases, reverting to its pre-event level after several days.”

Transaction fees on Bitcoin also spiked sharply. Value-weighted relative fees increased by about 127% above the pre-blackout average, or almost three standard deviations.

Despite this fee increase driven by greater demand for Bitcoin’s limited processing capacity, total net miner revenue still fell due to a decline in the number and total value of transactions.

Block rewards paid out to miners dropped by an estimated $118 million during the blackout, even as total fee income rose by around $44 million, translating to a net miner revenue loss of about $74 million over the event window.

Beyond Hashrate, The Market Impact of Mining Disruptions

By focusing on miners’ dependence on low electricity prices rather than on mining hardware or mining pools, the study has highlighted operational and geopolitical risks that can “adversely affect the whole blockchain network.”

However, it’s not just hashrate, transaction time, and transaction fees that are impacted until Bitcoin’s self-correcting mechanisms restore normal operation; the effect goes far beyond.

Besides the network, the study observed secondary market effects during the disruption. The researchers found that longer confirmation times coincided with higher short-term price volatility. Per the authors, Bitcoin’s return volatility rose more than that of ADA.

Moreover, exchange liquidity declined as bid-ask spreads widened and price differences grew across trading venues like Binance, Coinbase, Kraken, and Huobi. These changes indicate a breakdown in cross-exchange market integration and arbitrage efficiency.

Although these market impacts were significant, they were temporary. But they do show how infrastructure disruptions within mining can extend far beyond blockchain operations into broader cryptocurrency markets.

As the authors note, difficulty adjustment provides real long-run resilience, but only after the fact. It also occurs roughly every two weeks, and according to the authors, “a faster adjustment could potentially decrease the impact of such mining shocks.”

“In the long run, this resiliency attenuates the geopolitical and operational risks that stem from the geographical centralization of mining.” But in the short window before that adjustment kicks in, a concentrated mining base leaves the whole network exposed, subjecting the traders and exchanges that depend on it to “higher and more volatile fees in addition to opportunity costs due to slower settlement and missed gains from trade.”

It’s important to note that the study is not proof that Bitcoin remains dependent on Xinjiang or Chinese coal power today. Global mining has shifted substantially since the event, with the United States, Texas in particular, now accounting for a sizable share of global hashrate.

That said, structural vulnerability hasn’t gone away, though the specific geography of mining has shifted. This is evident in the ongoing mining pool concentration. Data shows that Foundry Digital (28.21%), AntPool (21.37%), and F2Pool (17.09%) collectively control 66.67% of the Bitcoin network’s hash rate.

But while mining pools are “considered a threat to decentralization,” they can involve many users who combine their resources and help enhance Bitcoin’s decentralization even in the presence of centralized mining pools.

What the research demonstrates is the underlying mechanism: whenever a large share of PoW mining capacity comes from a single regional grid, a correlated shock to that grid, whether a blackout, a natural disaster, a policy change, or a geopolitical disruption, can ripple far past hashrate numbers into higher fees, thinner liquidity, and a fragmented, less efficient market for the asset itself.

The most important conclusion here, however, is that the blackout did not threaten Bitcoin’s survival at all.

The protocol functioned as intended: miners continued validating transactions, blocks continued to be produced, and the network ultimately recovered without requiring centralized coordination.

But what the study also highlights is that while Bitcoin demonstrated strong protocol-level resilience, the physical infrastructure supporting mining remained susceptible to localized shocks due to geographic concentration.

The researchers therefore argue that improving the geographic distribution of mining operations would strengthen the overall resilience of proof-of-work networks while reducing the systemic risks associated with concentrated computing power.

For investors, this means miner resilience should be assessed through geographic footprint, grid exposure, power-source diversity, owned-versus-hosted infrastructure, and the ability to redeploy compute when regional energy conditions deteriorate.

Company in Focus: MARA Holdings

One of the oldest crypto companies, MARA Holdings was founded in 2010 but didn’t pivot to Bitcoin mining until many years later. Today, it is one of the largest publicly traded Bitcoin miners, with a market capitalization of $4.4 billion.

But with BTC price trading below its estimated production cost of roughly $70,000, the entire Bitcoin mining industry has been dealing with declining profits.

This pressure is visible in MARA’s Q1 2026 revenue, which dropped from $213.9 million in the prior year period to $174.6 million. The company posted a net loss of $1.3 billion, or $3.31 per diluted share, during this period, while its adjusted EBITDA was negative $1 billion. The company did retire approximately 33% of its outstanding debt, though.

In light of strained miner economics, MARA has been transforming into a company with increasing ownership of energy and infrastructure assets, focusing on expanding its infrastructure capacity into adjacent workloads.

Boasting approximately 1.9 gigawatts (GW) of total capacity, the digital infrastructure company is utilizing this infrastructure to satisfy the massive power demands of AI, high-performance computing (HPC), and critical information technology (IT).

This is being achieved through acquisitions of operational data centers, ownership of power infrastructure, and strategic partnerships to improve operational control and reduce mining costs. With this hybrid approach, the company can create a more scalable and resilient mining portfolio.

This includes a partnership with Starwood to develop capacity for hyperscale and enterprise tenants, a $1.5 billion acquisition of Long Ridge Energy & Power, and a 64% stake in European AI infrastructure company Exaion.

Most recently, MARA announced an agreement to acquire a large powered land plot in Matagorda County, Texas, intended for a computing campus that could host both AI data centers and Bitcoin mining machines side by side. The site spans over 1,200 acres and is expected to provide access to 1 GW of grid capacity by October 2027 and another 1 GW by April 2028.

All these moves have helped MARA shares rally (RLY ) 30% YTD, trading at $12 at the time of writing, while down 38% over the past year. It has an EPS (TTM) of -5.91 and a P/E (TTM) of -1.98.

When it comes to crypto, MARA still leverages Bitcoin as its core business, operating multiple mining sites that provide hosting services, including colocation and managed services, to institutional-scale crypto mining companies.

It actually maintains one of the largest corporate Bitcoin treasuries in the industry. With a stash of 36,303 BTC, MARA ranks 4th on the list of top 100 Public Bitcoin Treasury Companies, just below Metaplanet and Twenty One Capital (XXI ), which own 43,000 BTC and 43,515 BTC, respectively. The 1st place is captured by Strategy with its 843,775 BTC holdings.

By combining direct Bitcoin mining with opportunistic Bitcoin purchases, this “twin-turbo” approach allows MARA to grow its long-term BTC holdings through both production and acquisition.

In 1Q26, MARA mined a total of 2,247 BTC or 25 BTC per day but sold $1.5 billion worth of Bitcoin to fund the AI/HPC buildout. As for its mining economics, MARA’s hashrate in the quarter increased 33% from 1Q25 to 72.2 exahash per second while its cost per kilowatt hour was $0.04 for owned sites.

A striking move from MARA came from its strategic reorientation, with the company saying that it does not plan to make large-scale purchases of new ASIC mining hardware for the foreseeable future.

Against the backdrop of the blackout research, MARA’s initiatives represent more than simple business expansion. They reflect an industry-wide recognition that future competitiveness no longer depends only on deploying more mining hardware, but also on securing reliable energy supplies, diversifying infrastructure geographically, improving operational flexibility, and building resilience against economic risks.

This becomes especially important as Bitcoin rewards continue to get meager, all set to get halved to 1.5625 BTC per block in April 2028 when the 5th halving event takes place.

As Bitcoin mining becomes more industrialized, resilience will become just as important as raw hashrate in determining long-term success.

Conclusion

Bitcoin is the world’s most prominent decentralized digital currency, and as the latest study demonstrates, the protocol remains remarkably resilient even when a significant portion of global mining capacity suddenly disappears.

While the hashrate drop, fee spikes, thinner liquidity, and wider price gaps between exchanges reveal the network’s structural vulnerability, Bitcoin has been designed to tackle these very issues and eventually smooth things over.

But with the current concentration of hashrate among a handful of mining pools, alongside an industry-wide profitability squeeze pushing weaker miners offline, the underlying dynamics the study describes are still very much in play.

As Bitcoin adoption accelerates globally, the resilience of the mining network will increasingly depend on how effectively miners balance scale with decentralization, ensuring that the world’s largest cryptocurrency remains secure even when unexpected events put its infrastructure to the test.

Click here for a list of top bitcoin mining stocks.

References

1. Scharnowski, S. & Shi, Y. Bitcoin blackout: Proof-of-work and the risks of mining centralization. Journal of Financial Stability, 101569 (2026). https://doi.org/10.1016/j.jfs.2026.101569