Summary Even with the current BTC price appreciation, hashprice remains too low for the mining part of Bitfarms to be profitable. Miners are stuck in the endless loop of having to procure more efficient miners. Increased competition from countries with cheap electricity like Kazakhstan and Russia. Editor's note: Seeking Alpha is proud to welcome Emanuel Nemec as a new contributing analyst. You can become one too! Share your best investment idea by submitting your article for review to our editors. Get published, earn money, and unlock exclusive SA Premium access. Click here to find out more » This is my first coverage of Bitfarms (BITF), so I will not go into their story so far; there are other articles that explain it well. I want to focus on recent events from operations, the recently announced merger, and my opinion on how it will affect the company's future. Operations Bitfarms managed to increase their revenues from 41.5M in Q2 to 44.9M in Q3, while operating loss worsened from 23.7M to 39.4M in Q3. This clearly reflects a steep decrease in hashprice. Increases in revenues were mainly due to increased hashrate - it rose from 10.4 EH/s to 11.9 EH/s. Bitcoin's average hashprice between March 30th and June 30th was around 70 USD/PHS/DAY, between June 30th and September 30th, it was 45 USD/PHS/DAY, according to the hashrate index and the author's calculations. Between September 30th and December 27th, hashprice was around 53 USD/PHS/DAY, which will play an important role in Q4. But the question is, what will the hashprice be a year from now? To start, hashprice is based on four factors: BTC price, network difficulty, transaction fees, and block subsidies (currently it is at 3.125 BTC per block, and the next "halving" is projected for April 2028-so I will ignore it for now). BTC price rose 119% YTD (Chart 1); it is currently sitting at $92,800 per BTC. It is hard to say where the price will be in a year, but historically, it tended to rise significantly before making a pullback and consolidating for a year or two (Chart 2). I believe history will repeat itself, and BTC price will stabilize, make a pullback, and that will negatively impact the hashprice in the next year, maybe two, before BTC rises further. Chart 1 (Yahoo Finance) Chart 2 (Yahoo Finance) Second, the network difficulty rises when hashrate increases. Historically, it has been increasing steadily (Chart 3), and I surmise it will continue to do so, again negatively impacting the hashprice. Chart 3 - BTC Network Hash Rate (YCharts) Third are the transaction fees, which tend to spike when the BTC price is rising exponentially (compare Charts 2 and 4). Currently, they are at around 5% of the total block income, and without further spikes in BTC price or hype around events like runes or NFTs, I believe fees will remain under 10% of the block income, thus minimally affecting hashprice. Chart 4 - Transaction fees as % of total block income (Bitbo) From the points I stated above, we can surmise BTC hashrate will remain steady or will be lower in the foreseeable future (1-2 years), thus negatively impacting revenues. Since in Q4 hashprice was higher than in Q3 but lower than Q2, the company will probably improve net loss a bit in Q4, while remaining far from profitability. To achieve net profit, hashprice would need to increase dramatically, which I find unlikely as discussed above. This makes the business model currently unsustainable, which may be why the company is looking to expand its business into the HPC/AI territory, as I will discuss. Additionally, the company likes to state how total cash cost per BTC is lower than average revenue per BTC (Table 1-Q3 Earnings Report), while producing a negative gross margin of 26% for Q3 (Table 2-Q3 Earnings Report) and negative 25% for the nine months ended September 30th. Operating margins are looking even worse at -81% in Q3. Current cash cost may be lower, but cash cost doesn't include depreciation and amortization, which is included in the operating margin. To me, it looks like the company wants to redirect attention to other metrics that look more favorable. Table 1 (Q3 Earnings report) Table 2 (Source Q3 Earnings report) Unlike some other companies, notably CleanSpark (CLSK), Bitfarms is selling most of its BTC assets. In Q3 they sold 461 BTC, while adding only 41 BTC to the treasury for a total of 1188 BTC held in the treasury. The total value of said BTC is $109 million (calculated on December 27th-$92,000/BTC)-which is currently 14% of their market cap. CleanSpark has, for example, 29% of their market cap in BTC and is in the long term in a much better position to benefit from BTC price increases than Bitfarms. I also argue that Bitfarms has the lowest TTM operating margins compared to the other US-listed BTC miners, making it the least attractive (Chart 5). Chart 5 - Operating margins of BTC miners (Author's work based on data from Seeking Alpha) Planned upgrades and increases in mining efficiency Currently, the company is replacing Bitmain T21 Miners with Bitmain S21 Pro Miners. Bitfarms will be installing more powerful and efficient S21 Pro miners, operating at 234 TH/s and 15 w/TH, which represents a 20% improvement from the T21 miners in both energy efficiency and hashrate. CEO of Bitfarms Ben Gagnon talks about this in the interview with Bryce McNallie. He outlines the possibility of either improving energy efficiency or reducing electrical costs. Currently, they are mostly working on improving energy efficiency with the new miners while their cost of electricity rose. But in my opinion, so is the competition; bitcoin miner efficiency is constantly improving, with more advanced ASICs, while mining companies are stuck in the constant loop of having to replace old miners. Acquisition of Stronghold Digital Mining Bitfarms announced the acquisition of Stronghold Digital Mining (SDIG) in an all-stock transaction, with the goal of increasing its energy portfolio to over 950 MW by the end of 2025. Stronghold Digital has, in total, two waste coal power plants in Pennsylvania that are burning leftover coal from the coal miners. After the merger, they will have 65% of electricity generation in North America, while the rest is in Latin America. The company likes to offer the story of a transformative acquisition for a variety of reasons, from environmental leadership to power grid energy trading and HPC & AI. In my opinion, both power plants ( Scrubgrass plant and Panther Creek plant ) are relics of the past, not the datacenter power plants of the future, and Stronghold's operating loss proves it, as I will discuss. In my opinion, Stronghold Digital Mining has similar problems as Bitfarms, declining profitability (in the last quarter, a negative operating margin of 206%) and strong competition from countries with cheap and abundant energy resources like Kazakhstan and Russia ( legalized cryptocurrency mining in November 2024). After the merger, Bitfarms will do their best to improve margins and unlock synergies. Most likely, they will reduce SG&A expenses. But even if all of the Stronghold's Selling, General & Administrative expenses are absorbed into the current Bitfarms expenses, the Stronghold part of the merged company will still have a negative 103% margin. Can merging two unprofitable companies create synergies and improve margins? Yes. Do I believe it will make a meaningful difference? No. Competition from countries like Kazakhstan and Russia Russia and Kazakhstan have around 25% of the total global hashrate . Note that this data is before Russia officially legalized BTC mining in November 2024, and their market share is bound to increase in 2025, due to both cheap electricity and as a means to bypass western sanctions. Electricity in Russia currently costs $0.054 per kWH for residential customers and $0.067 for businesses; in Kazakhstan it costs $0.046 per kWh for residential customers and $0.074, while in the US the average is $0.169 per kWh. This makes BTC mining in the US much less profitable. HPC/AI opportunity The company states during the Q3 earnings call that they plan to disrupt the datacenter and AI industry with a hybrid AI/BTC mining/HPC combination. Currently, they are planning a 1-2 MW pilot site (Slide 1) to evaluate datacenter opportunity and possibly get some smaller contracts. I believe for BTC miners, there is a long-term opportunity in the HPC market, possibly much more profitable compared to BTC mining. Slide 1 (Company's presentation) One example in the industry would be Hut 8, a Canadian BTC miner, that in June 2023 signed an agreement with Canadian health authority Interior Health to provide colocation services from its Kelowna data center until 2028; in the last quarter they launched a GPU-as-a-Service business. Their gross margin from the HPC segment sits at 23% in Q3 2024. They didn't provide details about operating expenses for the specific segment, so it's hard to estimate the segment's profitability. I cannot deny there is an opportunity in the datacenter operations, but I believe it's still too early to put the money behind the company's plans; there are just too many unknowns. I'd prefer to wait for the results from the pilot plant and then do a reevaluation to see if there is indeed an opportunity. As soon as we have some results from the pilot plant and the water is clearer, I will provide a reevaluation. Risks to my theory The highest risk to my short theory would be continued BTC price appreciation and subsequently for the hashprice to increase. But for the reasons stated above, I find meaningfully higher hashprice unlikely. Second is the possibility of a short squeeze. The latest short interest data indicates short interest at 10%; for me, that indicates low risk of a short squeeze. Trading plan The company is currently at the YTD lows. I plan to see how it will react at this price; hopefully, we'll see a bounce toward $1.80-1.90, in which case I'd love to add to my short position. I plan to hold my short position until I see some data from their HPC pilot plant; success there could present an opportunity and trend reversal, but until that happens (mid to late 2025), I am bearish.
