Bitfarms Ltd. (NASDAQ:BITF) Q1 2024 Earnings Conference Call May 15, 2024 8:00 AM ET
Company Participants
Tracy Krumme - Senior Vice President, Investor Relations
Nico Bonta - Interim Chief Executive Officer
Ben Gagnon - Chief Mining Officer
Jeff Lucas - Chief Financial Officer
Conference Call Participants
Lucas Pipes - B. Riley Securities
Joe Flynn - Compass Point
Josh Siegler - Cantor Fitzgerald
Mike Grondahl - Northland Securities
Martin Toner - ATB Capital Markets
Operator
Greetings. Welcome to the Bitfarms' First Quarter 2024 Financial Results Conference Call. At this time all participants’ are in a listen-only mode. A question-and-answer session will follow the formal presentation. [Operator Instructions]
I will now turn the conference over to your host, Tracy Krumme. You may begin.
Tracy Krumme
Thank you. Good morning, everyone, and welcome to Bitfarms' first quarter, 2024 conference call. With me on the call today is Nico Bonta, Interim Chief Executive Officer; Jeff Lucas, Chief Financial Officer, and Ben Gagnon, Chief Mining Officer. Before we begin, please note that this call is being webcast with an accompanying presentation. Today's press release and our presentation can be accessed at our website, bitfarms.com. Under the investor section.
Turning to slide two. I'll remind everyone that certain forward-looking statements will be made during the call and that future results could differ from those implied in the statements. The forward-looking information is based on certain assumptions and is subject to risks and uncertainties and I invite you to consult bit Bitfarms MD&A for a complete list. Please note that references will be made to certain measures not recognized under IFRS and therefore may not be comparable to similar measures presented by other companies.
We invite listeners to refer to today's press release and our MD&A for definitions of the aforementioned non-IFRS measures and their reconciliations to IFRS measures. Please note that all financial references are denominated in U.S. dollars, unless otherwise noted. I would also like to add that we will be attending the following upcoming equity conferences. The B. Riley Securities Institutional Investor conference in Beverly Hills, California on May 22 and May 23 and the Northland Growth Conference, which is virtual on June 25. If anyone is interested in meeting with us on those dates, please reach out to me or a sales representative from those firms.
And now it is my pleasure to turn the call over to Nico Bonta. Chairman cofounder and interim CEO Nico. Please go ahead.
Nico Bonta
Thank you, Tracy. And thank you, everyone, for joining us today. For those of you who might not know me, I founded the company along with Board Director Emiliano Grodzki in 2017 and have served as chairman of the Board since. As you will recall, on March 25, we announced a CEO transition whereby Jeff Murphy would remain CEO until we concluded our executive search. However, his departure has now been expedited due to a legal claim against the company brought by Jeff on Friday, May 10. This also caused us to reschedule our earning calls in order to provide time to incorporate the required subsequent events into our filings.
Jeff claimed included damages for breach of contract, wrong dismissal and aggregated and punitive damages in the amount of $27 million. The company believes the claims are without merit and intends to defend itself vigorously. As a result, Jeff's termination was accelerated, and I stepped as interim CEO until the conclusion of our CEO search which we expect to occur in the next several weeks.
Importantly, I would like to emphasize my confidence in the strong leadership team we have built, the significant opportunities ahead of us and dedication of our employees to meet our growth target decision. We expect minimal disruption from the CEO transition. Our 2024 growth plans will not be impacted, we are moving full steam and ahead and expect to achieve our previously stated guidance of 12 exahash in Q2 and 21 exahash in Q4.
Now turning to slide four. I am pleased to turn the call over to Ben, our chief mining officer who will discuss our Q1 results, updates and our growth plan and exciting outlooks for the year ahead.
Ben Gagnon
Thanks, Nico. Turning to slide five. Since we announced our transformative fleet upgrade in November and additional minor purchases in March investor interest has increased. Before we dive into our Q1 results, I would like to take a minute to provide a brief company overview for any newcomers. Bitfarms is a vertically integrated global Bitcoin mining company that provides investors with high-quality exposure to Bitcoin. We measure our success on three key performance indicators, hash rate, energy efficiency and direct cost to mine known as hash cost.
This year, Bitfarms is poised to deliver the greatest hash rate growth and cost improvement in our history with industry-leading benchmarks. We expect to reach 21 exahash, nearly tripling our current hash rate and nearly doubling our operating capacity this year. With the development of our new PASO PE site, we now have 12 farms spread across North and South America with an additional farm and development in Paraguay. We will continue to opportunistically evaluate new geographies, new markets, and new opportunities.
In Q1, 2024. Our revenue and margins continue to improve as Bitcoin prices moved upwards. Total revenue of $50 million increased 67% year-over-year and 9% over Q4 and our adjusted EBITDA increased 50% to $21 million versus Q4 last year. In addition, due to our success recouping our Canadian VAT, we will soon receive a $24 million cash refund that will be used to fund our growth.
Turning to slide six. We are on track to achieve our 2024 guidance 21 exahash a 223% increase and 21 watts per terahash a 40% improvement while growing our operational infrastructure by nearly 80% to 428 megawatts. We are front-loading the majority of energy efficiency improvements and growth in the first-half of this year with 70% of our new miners being allocated to upgrade our existing mining farms. This will ensure the fastest improvement in fleetwide energy efficiency by replacing the least efficient miners first. The remaining 30% of miners will be energized at new locations, primarily in Paraguay.
With lessons learned from the past halving over the last two years, we deliberately deleveraged our balance sheet and followed a strategy emphasizing low cost, vertically integrated operations. Now, with no debt, a rapidly increasing hash rate, improving energy efficiency and cost reductions on a per-unit basis we are aggressively entering this new era to capture as much upside as possible while maximizing returns and driving long-term value for shareholders.
Turning to slide seven. Our growth this year is a major inflection point for the company and represents a golden opportunity for current and prospective investors. The strategic timing of this inflection point should not be overlooked. So why are we growing so aggressively right now? In 2023. We analyzed over six years of purchase history for every miner, evaluating the following the price we paid, the time we plugged them in, how much revenue was generated, the cost and profit, what the payback was, and the cost-effectiveness of each purchase. What we found is that the most important factor for determining a good miner investment is timing. As timing ultimately determined costs.
The miners we purchased in the months leading into and out of the last halving in 2020 were incredibly cost effective and paid themselves back between five and eight times over, delivering by far the best returns on investments. Emboldened by the data, we launched our transformative fleet upgrade strategically timed with these cycles in mind. First, we secured low-cost and stable power of 170 megawatts in a region where we could build out quickly, deploying the remaining miners expeditiously and predictably. Second, we created a plan to quickly and cost effectively upgrade every single one of our existing farms by upgrading our older and less efficient miners.
Third, in November 2023 with Bitcoin around $37,000 and the halving just months away, we jumped on an announcement by Bitmain regarding a new series of more powerful and more productive miners. We secured some of the lowest prices seen in years and negotiated a miner option, an industry first. We moved quickly and launched our fleet upgrade plan, and in March followed up with even more miner purchases, purchasing nearly an additional 24,000 of the best-performing machines on the market with competitive pricing.
Lastly, in January 2024, we purchased land for Yguazu Paraguay Farm. This land is strategically located across the road from a recently constructed ANDE substation which is fed by a 500-kilovolt line directly from the Itaipu hydro dam and has additional acreage to accommodate further expansion. In summary, this transformative plan speaks to our discipline by first securing megawatts with a plan for every miner purchased and are strategically timed investments by purchasing miners at attractive prices.
Turning to slide eight. By executing on our growth plan, we are well-positioned to gain market share, and we believe our planned growth towards our year end portfolio represents the best opportunity on market. By the end of 2024, we will have increased our hash rate 223% from 6.5. to 21 exahash. Two, increase our energy capacity nearly 80% from 240 megawatts to 428 megawatts. Three, improved our energy efficiency 40% from 35 to 21 watts per terahash.
Four, increase the total miners deployed by 48% and five increase our total farms under management by 18%. Combined, we believe these figures make up the most meaningful and concrete growth plan announced among the publicly traded miners. As a result, we will close the year with a geographically well-balanced portfolio of nearly 100,000 highly efficient and competitive miners purchased at some of the lowest cost seen in years, operated at 13 farms spread across North and South America, and primarily powered with consistent and sustainably low cost hydropower.
Turning to slide nine. Based on our 2024 growth, we will be improving and expanding faster than the Bitcoin network as demonstrated by these two sensitivity tables. On the left, we show our growth relative to the network hash rate. By growing faster than the network, we will capture market share and increase the numbers of Bitcoins earned per day. Our 223% hash rate growth to 21 exahash will be a significant revenue driver and spread our G&A over a larger number of Bitcoins.
On the right, we show how our improvement in energy efficiency are also expected to be faster than network growth. This, combined with our largely stable and competitive energy rates, driven by approximately 80% surplus hydropower should result in increasingly lower direct energy costs per Bitcoin mined over the year.
It is important to note that these results are determined by our continued improvement across key performance indicators of hash rate and energy efficiency relative to network hash rate. They are not reliant on Bitcoin pricing. As the widely anticipated bull market takes off in 2024, these improvements will ensure we are well-positioned to capture upside with quickly expanding mining margins. This is how we plan to outperform the industry this year.
Turning to slide 10. In Quebec, we are executing. Our transformative fleet upgrade, generating a low-risk and stable growth pathway. Quebec boasts competitive rates on hydropower and cost-effective growth as we upgrade our state-of-the-art facilities with our new T21 miners. We have already completed two farm upgrades, generating a remarkable 51% improvement in energy efficiency and illustrating the transformative impact of our fleet upgrade. With three more farm upgrades underway, our total hash rate and energy efficiency in Quebec will improve dramatically throughout Q2.
Additionally, we have begun testing recycled heat from hydro mining technology with the deployment of 153s plus miners. We are incredibly excited about the potential of this technology for heat recapture and reuse. The province of Quebec has tremendous demands for residential, commercial and industrial heat, and we believe the hydro miners are potentially the best technology available to expand our business in the province, drive down costs, and help achieve the province's climate goals by recycling our primary waste product, heat.
Turning to slide 11. Argentina remains a low-cost jurisdiction with significant expansion possibilities. In September, we will be deploying up to 6400 new T21 miners that will improve our site's energy efficiency from 32 watts per terahash to 23 watts per terahash while increasing the site hash rate by approximately 850 petahash.
Turning to slide 12. Now let's turn our attention to Washington, currently our only site in the US. With no curtailment and incredibly high uptimes it is also one of our most productive sites. We are currently expanding and doing site modifications to accommodate additional T21 miners to be deployed in October this year. This deployment will improve site efficiency approximately 27%, from 30 watts per terahash to 22 watts per terahash and increased total hash rate approximately 46% to 910 petahash. Notably, the US represents the smallest region in our portfolio and we are actively looking at numerous opportunities to increase our exposure in the US in both 2024 and 2025.
Turning to slide 13. Let's talk about our expansion in Paraguay, which represents the largest growth in our plan this year. With megawatt expansion of 170 megawatts and 9.2 exahash growth from our two new farms, we have already installed thousands of miners in PASO PE and are finalizing work on the substation.
When complete, PASO PE will add over 3 exahash and 70 megawatts to our portfolio growing our megawatts under management 29% to 310. Construction at our third site in Paraguay, Yguazu, is well underway and on track to be energized in Q4, contributing approximately 5 exahash and 100 megawatts with energy efficiency of 20 watts per terahash in 2024.
Paraguay's economic and political landscapes make it highly attractive. The majority of its political leaders and senior leadership of ANDE, the state-owned power distribution company, recognize the tremendous economic benefits Bitcoin mining operations bring and are very supportive of legal Bitcoin mining ventures. Numerous senators and public officials sign letters and commented publicly about the positive outcomes of Bitcoin mining. These include increased employment opportunities, capital spending, higher revenues from energy sales for legal Bitcoin mining operations and increased electrical infrastructure investments.
Importantly, Paraguay boasts a massive surplus of energy relative to local demand. Over 50% of its power is exported to neighboring countries for a fraction of the price legal Bitcoin mining operations pay. Bitcoin mining represents a meaningful conduit to sell excess renewable energy to a global marketplace and drive significant economic growth. In turn, we benefit from access to low-cost, reliable, sustainable green power with almost no curtailment, a skilled and cost-effective labor pool, expedited construction schedules and a business-friendly environment supportive of our industry.
Building upon our success in Paraguay, we recently signed an agreement with the state-owned utility ANDE, doubling the energy capacity of our Yguazu site with an additional 100 megawatts in 2025. Growing Yguazu to 200 megawatts increases our 2025 megawatts under management 23% from 428 megawatts. To 528 megawatts. Importantly, this expansion takes advantage of our existing construction plan, amortizing costs over a greater amount of infrastructure and driving down overall cost per megawatt.
Changes to construction plans and equipment orders are already in progress, and we are currently analyzing potential deployment plans. To contextualize the impact of this assuming a similar miner model and the same 20 watts per terahash efficiency already planned for Yguazu, this additional 100 megawatts could support an additional 5 exahash in 2025.
Turning to slide 14. I'll now hand the call over to Jeff Lucas to talk about our financial performance.
Jeff Lucas
Thank you, Ben. Turning now to slide 15. This is a great time for the industry, and we have been preparing for the post-halving world with a strategic growth plan, an accretive financing strategy and a debt-free balance sheet. With our funds from operations along with our potential tax refund and the judicious use of our ATM financing facility, we are well positioned to grow our capacity from 240 megawatts to 420 megawatts this year and hit our year-end 21 exahash per second target.
Here are a few first-quarter highlights large in comparison to the fourth quarter of 2023. Revenue of $50 million was up 9% over the fourth quarter and up 67% year-over-year. The quarter-over-quarter comparison reflects a 44% higher average Bitcoin price offset by 24% pure Bitcoin earned during the first quarter.
We earned 943 Bitcoin in the quarter and paid to 1236 Bitcoin in the prior quarter. Primarily the result of an increase in average network difficulty of 21%. Mining revenue with $49 million, compared to $45 million in the prior quarter. Gross mining profit was $29 million, or 59% of mining revenue, compared to $23 million or 52% of mining revenue in the fourth quarter.
General and administrative or G&A expense was $13.2 million, including $3.1 million of noncash compensation expense and approximately $1.6 million of nonrecurring severance charges associated with the management change announced in March. Net of these items. Cash G&A expense in the first quarter was $8.5 million down 11% from the previous quarter. Our operating loss was $24 million in comparison to a $13 million loss in the fourth quarter. The first quarter operating loss included a $39 million depreciation expense in comparison to $22 million in the prior quarter. First quarter depreciation expense included $18.5 million of accelerated depreciation expense associated with the older miners we replaced by the new T21 miners in Quebec.
Under the Farnham program, the existing miners are being depreciated on an accelerated basis over the remainder of their expected operating life. As such, a higher level depreciation expense is expected over the remaining quarters of 2024 as the new miners are brought online.
In the first quarter, we recorded a $9 million noncash gain for the revaluation of financial liability for warrants issued in earlier finances, compared to a $38 million noncash charge for the revaluation of this financial liability in the fourth quarter. Under IFRS we are required to recognize the liability for these warrants, even though they cannot be settled for cash. First quarter net loss was $6 million, a loss of $0.02 per share in comparison to a net loss in the fourth quarter 2023 of $57.2 million or a loss per share of $0.19.
Now let's turn our attention to operating performance in per Bitcoin metrics. Our corporate cost of electricity during the quarter was $0.41 per kilowatt hour. Substantially unchanged from the prior quarter. Due to Canadian tax legislation proposed in February of 2022 our direct cost of production since that date has included a 15% value-added tax or VAT on Canadian energy costs.
While this continued to impact our direct cost through the first quarter, I'm delighted to report that we succeeded in obtaining approval from the Canadian Tax Authority against the VAT regulations. Going forward we will be able to recover the VAT and not have to reflect the tax on electricity costs. This is a significant reduction in our operating cost.
Were we not have included the VAT in our first quarter electricity rate, our corporate cost electricity would have been $0.37 per kilowatt hour. Furthermore, this VAT recovery will entitle us to a $24 million refund for the tax amounts paid since February 2022, which will be applied towards funding our 2024 growth initiatives. In the first quarter, our direct cost of production for Bitcoin was $20,500. Overall, excluding the Canadian VAT, our corporate direct cost would have been $18,400, $2,100 lower than our actual corporate direct cost.
Turning now to slide 16. Adjusted EBITDA increased to $21 million of 50% from the fourth quarter. This equates to cash profitability per Bitcoin of $22,700 or about double the $11,200 per Bitcoin in the fourth quarter. Adjusted EBITDA margin increased to 42% from 30% in the previous quarter.
I want to point out that our adjusted EBITDA is a very straightforward measure comprised simply of our cash profit per Bitcoin times the number of Bitcoin miners, plus a profit earned on a Volta subsidiary.
As an IFRS filer, we do not mark to market our Bitcoin holdings and we do not include this or any other balance sheet valuation changes in our adjusted EBITDA. Adjusted EBITDA for us is purely a measure of the cash profitability of our mining operations and the modest profit contribution of Volta Electric subsidiary.
Turning to slide 17. At March 31, we held cash of $66 million in Bitcoin, valued at $58 million for total liquidity of $124 million. This compares to $118 million of total liquidity at December 31. On March 11, we commenced our ATM program and raised $38 million in net proceeds during the quarter, which are earmarked to fund our growth initiatives in our fleet upgrade.
In addition, we receive $1.7 million in net proceeds from the sale-leaseback of our Garlock facility. From March 31 to May 14, we raised an additional $83 million under the ATM program to further fund our growth.
During the quarter, we paid off the remaining equipment financing debt, leaving us debt-free at March 31. I'll point out, however, that we show a $1.6 million of long-term debt on our March 31 balance sheet to reflect the IFRS accounting associated with the Garlock sale-leaseback. Importantly, I want to underscore that we have sufficient liquidity to pay for all the miners needed to reach 21 exahash per second this year.
Turning now to slide 18. Before we open the call for the questions, I'd like to summarize here. We are dramatically offering our operating profile via our ongoing upgrade and expansion plan. Recent miner upgrades are already delivering major efficiency gains, and further gains throughout the year should contribute to post-halving margin improvements.
Driving growth and improving our portfolio, we are on track to achieve 21 exahash per second and 21 watts per terahash efficiency in 2024. With an industry-leading Bitcoin mined per exahash Bitfarms distinguishes itself through exceptional margin performance demonstrating operational efficiency and profitability in a highly competitive industry. This preparation is underpinned by our robust balance sheet and strong liquidity, which are crucial for sustaining growth and capitalizing on new opportunities.
With a strong leadership team, with a proven track record of driving profitable growth, our operational excellence and strategic vision have been instrumental in the success over the past six years, including two halving events. Furthermore, Bitfarms' commitment to ESG reflects our dedication to sustainable and responsible mining practices. It's gratifying that our largest projects now under development will draw power from the Itaipu dam, the third largest hydropower power facility in the world. Overall, we are very well-positioned for continued growth and success in 2024 and beyond.
With that, I'd like to turn the call back to the operator to open this call for questions.
Question-and-Answer Session
Operator
Certainly. At this time, we will be conducting a question-and-answer session. [Operator Instructions] One moment, please, while we poll for questions. Your first question for today is from Lucas Pipes with B. Riley Securities.
Lucas Pipes
Thank you very much, operator. Good morning, everyone.
Ben Gagnon
Good morning, Lucas.
Lucas Pipes
Jeff. You just mentioned there in your remarks that all the miner costs are kind of covered with the liquidity you have. So in that context, how should we think about the use of the ATM over the balance of this year? Do you expect to raise capital for infrastructure or other reasons? Just curious how you think about that. Thank you very much.
Jeff Lucas
Sure. So let me, I'll be glad to address your question here. First of all, we have been active with the ATM actually since March 31, and the liquidity that we have today is higher than what we reported on March 31. The other point to keep in mind here as we pointed out, both Ben and I during the call is that we are receiving a tax refund, which we expect to have by the end of June for roughly $24 million here. So that addresses in large part both requirements that we have for our miners as well as the infrastructure that we have going forward.
All that being said, we are looking to continue to judiciously use the ATM going forward. And our conjecture, to be candid over the balance of the year here we can see ourselves using $50 million or a little north of that. But again, we're being very careful in terms of how we manage our ATM here going forward.
One thing I also want to share with you is that when we sit here with our projections here, including our anticipated use of the ATM here going forward one thing that we have not reflected in our numbers here is that we are projecting here that excess cash flow from operations on a monthly basis, anywhere from about $6 million to $10 million over the next several months. Now, of course, that's dictated in large part in terms of what happens with the price of BTC.
So while we have that in our pocket, we'll obviously avail ourselves of that, particularly through our Synthetic HODL program. But that aside, we have not factored into some of the numbers I just mentioned here with the plans for the ATM. If I can take advantage of this, Lucas, one other comment here in terms of the benefit of the ATM to us here.
So when you think about what we're doing here in terms of our growth, for every hash rate that we're adding here, it's costing us between $21 million to $23 million all in. That includes infrastructure, miners and some logistical costs associated with getting those miners up and running at the various locations in Canada and also South America as well here.
You look at valuations for mining companies here and generally you're seeing in the range of about $45 million per exahash. So in our mind here, while we are being very careful and very thoughtful with the ATM, we are certainly husbanding our money very carefully here and our funds here. We do indeed see the use of it certainly validating the accretive nature in what we're doing here.
Lucas Pipes
Thank you. Thank you very much for those comments. On the power cost side, you're adding capacity of power and I wondered if you could kind of walk us through the impact as maybe your mix changes a bit. Thank you very much.
Jeff Lucas
Ben do you want to speak to that, or do you like to just talk about some of the economic impact we anticipate going forward?
Ben Gagnon
Sure, I can speak to that. Really at a high level, Lucas, the cost that Jeff just put out there for Q1 was about $0.41. The contract that we have for additional power that's under construction in Paraguay is a $0.39 contract which is fixed with no annual inflation adjustment mechanisms in that for the coming years. And when you add VAT on top of that, that comes out to, I think, just approximately about $0.42.
So the average cost for our power with what we had last quarter and just looking at where the all-in cost for power is in Paraguay, it doesn't really meaningfully change the average price for power across our portfolio. It just gives us a much greater amount of power in our portfolio for about the same cost.
Lucas Pipes
Got it. Okay. No, that's very helpful. I appreciate all the color and continue. Best of luck. Thank you.
Ben Gagnon
Thank you.
Jeff Lucas
Thank you, Lucas.
Operator
Your next question is from Joe Flynn with Compass Point.
Jeff Lucas
Hello, Joe.
Joe Flynn
Hi, guys. Thanks for the question. I had a question regarding you guys mentioned potential expansion opportunities in the US. Ultimately, if you could provide more color on what that potential deal pipeline looks like, and I'd be kind of curious to get your perspective on as you're competing for power with in demand HPC is ultimately do you think Bitcoin mining is moving more internationally? Thanks.
Ben Gagnon
Thanks, Joe. We're looking at multiple deals right now. We have a global view on power and getting cost-effective megawatts at scale in various jurisdictions where the economics are compelling. We think the US has a lot of very good opportunities for 2024 and 2025 expansion. And we are evaluating those right now. I think the good thing and the added benefit to our existing growth plan here is that we've got a lot of built-in extra capacity here with the miners that we've purchased.
The growth to 21 exahash basically is using a much lower configuration for those miners, and they're capable of pushing through. So if we do find additional megawatts, ideally in the United States or somewhere else, with compelling economics we'll be able to grow our hash rate using the miners we already have on hand that we already have the liquidity to pay for to reset 21 exahash. Getting more miners or getting more hash rate out of those miners and driving even greater hash rate growth.
So those are the kind of things that we're looking at right now. Are there good opportunities for us to execute on maybe some additional megawatts, which would enable us to get even greater utilization out of our miners and our growth plan for this year? As we move forward, obviously, we're communicating those deals to the public. But we do have numerous different deals in our pipeline right now in the United States and otherwise.
Joe Flynn
Great. Thanks. And then just on the Bitcoin treasury management side, going forward, should we expect continued sales of majority of your production? And if you maybe can comment and provide more color on just you guy's option exposure and how that is done here with Bitcoin up strongly year-to-date.
Ben Gagnon
Sure. Actually glad to do that here. We do anticipate some increases in our HODL. It may not be all the expenses. In large part because we put in place here what we've talked about in the past called the Synthetic HODL, which sort of in our mind gives us the best of both worlds. Just for clarification in terms of how we treat this here. What we normally do is when we excess cash flow from operations, we will generally sell those Bitcoin.
And by way of example, here we may take 90% of that Bitcoin proceeds and use it to fund our capex going forward which of course means there's less having to tap into the ATM, the other 10% or 15% of the proceeds on the Bitcoin we used to buy long-dated call options here. So in essence here, while the actual HODL itself will be increasing to a modest degree here, our exposure to the upside of Bitcoin will be that much greater by virtue of Synthetic HODL.
And I'll just add a couple numbers here, that's actually worked out very well for us. While we don't normally share what the actual achievements of our gain share that we realized here in our Synthetic HODL, but in the fourth quarter of last year we actually had a gain about $1.6 million, and it was up about 119% return.
For the first quarter of this year, we actually had a unrealized gain about $3 million and roughly over 500% return here. So this has actually worked out very handsomely for us. It, again, provided us with probably the cheapest source of capital out there versus having to access the capital markets here and also still gives us a preservation of the upside or a Bitcoin here. Does that answer your question fully there, Joe?
Joe Flynn
It does, yes. Thanks. Appreciate all the color. That's all for me.
Ben Gagnon
Sure.
Operator
Your next question for today is from Josh Siegler with Cantor Fitzgerald.
Josh Siegler
Yes. Hi guys. Good morning for taking my question. My first one I just want to follow-up on Joe's question, which is just now that we're post-halving, have you seen any difference in terms of deal terms becoming more attractive from peers that might be more squeezed? Just kind of curious have the types of deals coming to your desk changed at all over the past month or so?
Ben Gagnon
Yes, thanks for the question. I think for the most part, we haven't seen the full impact from halving just yet, having just happened not even a full month ago as of today. And it still takes miners usually a few weeks or a few months in order for those lags between a changing market condition and really feeling any sort of a need to make any sort of changes in their operation or in their structure. So I don't think the deals have changed at all over the last couple of months, more or less it's still the same.
Hash price of $0.05 is the level that we anticipated where there would start to be some level of response from the market whether that be slower growth or under clocking of machines or de-racking older miners. We did expect those kind of pressures to start happening at the $0.05 level. So with that hash price basically there, right now, I don't think there's a whole lot of change here happening at internal operations.
But certainly as hash price kind of stabilizes and we find out where we're heading here that will be a big driver if it goes down or if it goes up, there'll be some definite changes to the potential opportunities out there.
Josh Siegler
Yes. Got it. That makes sense. And for my follow up, I was wondering if you could elaborate a little bit more just on the cost of infrastructure build out in Paraguay versus your other sites.
Ben Gagnon
Sure. Jeff, do you want to handle that one?
Jeff Lucas
Sure. Let me start off and then you can add a little more color Ben as you see fit here. So actually our cost in the Paraguay is generally a little more attractive actually we're finding than it is in other parts of the world here. And we target here to do under $350,000 per megawatts build out here, and the rates actually that we are incurring are the cost that we're actually incurring in Paraguay are less than that, actually more around the $300,000 per megawatt range here. So to us that's pretty good, always looking for opportunities for improvement here. But that certainly falls within our guidelines and ROI and our payback projections here.
Ben Gagnon
Maybe to just add one quick comment there. All the developments there in Paraguay are greenfield developments. This is brand new sites that we're developing from the ground up, including the substation, all the interconnections, the actual facilities themselves, and of course building out the mining infrastructure and the support systems in order to make all of this work. And we're doing this in a rather accelerated timeframe. The PASO PE site is going to be done in well under a year, and same thing with the Yguazu site from start to finish.
So we've got very aggressive construction schedules and a very talented labor force that we're drawing from there, which helps to make sure that the deployments are very fast and we get the most out of unexpected bull run later in 2024. And it also helps to drive down the cost when things just don't take as long.
Jeff Lucas
And by the way, if I can just add one additional comment here, because this is actually disclosed in our financial statements. We should talk about as a subsequent event, the additional 100 megawatts that we're getting in Yguazu and Paraguay, and that's for 2025, just to be very clear here. And our estimate of the actual infrastructure cost to build that out is roughly $23 to $25 million.
The point to be garnered from that is the fact that while we are looking at costs of roughly 300,000 or little north of 300,000 per megawatt for Paraguay here, the marginal cost of additional build-out at those existing locations here is actually lower. And in this case for next year for the additional a 100 we're talking, we think around $230,000 to $250,000 per megawatt.
Josh Siegler
Okay, great. Thanks. Appreciate the call.
Operator
Your next question for today is from Mike Grondahl with Northland Securities.
Mike Grondahl
Hey guys, thanks. Could you talk a little bit about Argentina and how high that is on your list for future expansion?
Jeff Lucas
Well, we are asking…
Ben Gagnon
Do you want me to jump in?
Jeff Lucas
Yes, why don't you go ahead and I'll follow-up?
Ben Gagnon
Go ahead.
Jeff Lucas
Okay. So first of all, Argentina is very attractive for us. I think we pointed out here that we've actually, during the lower energy cost summer months for Argentina entered into a six month contract at $0.21 per kilowatt hour. Now that is going to be higher for the five months that follow the winter months, which are coming up shortly. And we are anticipating higher energy costs in Argentina, for the next five months or so here.
But we've mentioned in the past, and we're holding to it that the average year round cost of energy in Argentina is roughly three to three and a half cents here. We've got about 54 megawatts in now. We have the opportunity, as you may recall, to go up to 210 megawatts here. We've actually, we're looking at that carefully. We're pleased to see things sort of settling themselves out economically and politically what's going on in Argentina, but we were sort of on hold for a while.
While those things didn't settle out because really vis-a-vis some of the other opportunities that we had available to us, including Paraguay, they were just more attractive when you do sort of a risk adjusted rate of return, which is how we look at our projects here overall. So while we are optimistic and we see further upside with Argentina we are moving that cautiously. Right now we're really focused on other opportunities, particularly in Paraguay. Ben, do you want to add to that?
Ben Gagnon
No, I think you handled it well, Jeff. I think that's exactly the point. We're just going to take advantage of our best opportunities right now with Paraguay and we've got plenty of optionality there with Argentina as the condition settle and we have a better outlook on the future or a more clear outlook on the future.
Mike Grondahl
Got it. That's helpful. And, hey, secondly, can you guys kind of talk to a little bit what you're looking for in a new CEO and what gives you the confidence that you think you can have an announcement in? I think somebody said Nicolas, maybe a few weeks.
Jeff Lucas
Nicolas, do you want to try that or would you like us to step in for you? I know you're far away. Let me step in because we may have connection issues here with Nico here. So, Jeff was very helpful for us in taking our company to the next level as we matured. We were very entrepreneurial, very creative company initially. He helped put in place these sort of organizational and professional infrastructure that really allows us to be a scalable organization on a global basis.
When Jeff first came on board, we were in one country, Canada, we are now in four and most importantly we are positioned to look at opportunities up and down the North and Southern hemispheres. So we are very well positioned now that we are well positioned and we have that scalable organization effect here, management and the board are very focused now on where the growth opportunities are going forward. How do we capitalize on this administrative and organizational infrastructure we have in place here? And that's how we really take advantage of it.
And there's a wealth of opportunities within mining and to a degree outside of mining that we really now want to begin pursuing more aggressively here. We've got a great team with Ben and others that really play a role in identifying terrific opportunities here for us to really improve our shareholder value here. And we are now seeking a CEO who we can really step in and help drive that process going forward.
Mike Grondahl
Got it. And then, hey, one last one, Jeff. Have you guys provided any updated thoughts about how Bitfarm kind of views the HPC AI opportunity? Just be interested to hear any updated thoughts there.
Jeff Lucas
Well, outside the economics, I'm going to pass that one over to Ben.
Ben Gagnon
Sure. I'm happy to address that. You know, obviously we've taken a look at the HPC and the AI opportunities, but when you look at our business and where our core competencies are, really it's in building and operating the world's best Bitcoin mining infrastructure. And when you look at what Bitcoin mining infrastructure is versus HBC and AI infrastructure, there really are not a whole lot of similarities other than the fact that they both have large demands for power. For us, we have very, very competitive costs when we're building out our Bitcoin mining infrastructure and the few hundreds of thousands of dollars per megawatt.
For us to do HPC or AI infrastructure, really, it's measured in the millions of dollars per megawatt. And those numbers vary quite widely depending on the quality and the location and everything in the data center. But you're seeing numbers anywhere, honestly from about $4 million to $12 million a megawatt. For the infrastructure side, the higher amounts of capex is also similar on the computation equipment itself. So when you're looking at the relative cost of the GPUs per megawatt and other supporting equipment relative to let's say Bitcoin mining equipment per megawatt, the costs have a very similar kind of overweight component here where we were talking about tens of millions of dollars per megawatt as opposed to a few million.
So the costs are very different. The businesses are very different and I think that for investors who are looking for HPC and AI exposure, I think it's probably better from the investor's perspective for them to manage that in their portfolio directly. If they want high quality Bitcoin mining infrastructure exposure, that's what we provide through our Bitfarm shares.
And then if they want to have that HPC AI, it's probably better to put that through a separate vehicle and just control that allocation in their portfolio directly. I think that's more what the investors and the market is searching for. So kind of a high-level response on that.
Mike Grondahl
Sure. No, hey…
Jeff Lucas
Ben answered that pretty aptly, Mike. Let me just add a couple more points here. We take great pride in being the best operators in the business and we're convinced that we could operate a high-performance computing center very, very effectively. But first of all, to be very direct here and very straightforward, our cost of capital, while we manage it very carefully and it's attractive, it's not attractive with some of the larger players out there.
And for us to have to make the investment to be effective in high-performance computing here, as Ben pointed out here, doesn't really make the best sense for our shareholders versus doing what we actually do very well here in terms of very efficiently running mining operations here.
The third point I'll point out to you here is that as we explore the lowest cost structure we can get out there, we always look at power opportunities, which are a different nature in high-performance computing. We are in a position that we can take advantage of lower-cost power, recognizing that there can be some curtailment or interruptability to it. High-performance computing cannot afford that.
So for us and for what we do very well here, it just makes a lot more sense really for many accounts for us to really focus more on mining and not to at this juncture really get sidetracked by high-performance computing.
Mike Grondahl
Fair. Hey, good to hear your perspective. And good luck the rest of '24.
Jeff Lucas
Thanks, Mike.
Operator
Your next question for today is from Martin Toner with ATB Capital Markets.
Martin Toner
Good morning. Thank you for taking my question. I think investors would benefit from a little more detail on timing of capex through end of 2024. Can you give us a little -- talk through that a little bit for us and maybe include how much of the difference in exahash from now till 21 at the end of 2024 has already been paid for.
Jeff Lucas
Sure. Glad to talk to. So first of all, importantly, as we pointed out during the other part of -- the earlier part of the call here, we have covered actually the cost, the remaining cost that we have for the miners to get it to the 21 exahash going forward. We do have additional spending of course, that we have to do for the infrastructure here. And that actually spoke to the first question that Lucas asked at the beginning of the call here.
So to put your first in very concrete number here, if you take a look at our financial statements, we have a section called commitments. And on that there's a specific section saying what's left to pay for these miners. And it indicates $156 million of which roughly here, $40 million is in the second quarter, $85 million is in the third quarter, and roughly about $30 million is in the fourth quarter.
The good news here is that about $40 million that we actually have in the second quarter almost all that has already been paid here. So we're in pretty good shape here. Where we do see probably the lion's of our payments going forward, as evidenced by that commitment schedule here is really going to be in the third quarter.
So when we are anticipating to point out roughly that $86 million here going forward, and then also the infrastructure costs particularly what's going on in Yguazu is really going to be felt in that quarter. So we've actually got -- we're anticipating roughly $50 million build-out remaining for the infrastructure, and I think most of that's actually going to be happening probably in the third quarter, Martin.
Martin Toner
That's fantastic. Philosophically, and I know it depends on the nature of the infrastructure, but from start to finish, where are the costs incurred and over what time period? Can you kind of give people just like a bit of a rule of thumb for the types of mines that you are building?
Jeff Lucas
I'm sorry. I'm probably not fully understanding your question as a…
Martin Toner
Just like from start to finish are -- like it takes, let's say start to finish takes a year and a half with a quarter of the capex spent in the first half and then, what percentage is spent in the end. Just wondering if there's a simple rule of fund that you guys think about that investors can use to think about what kind of expect -- like what kind of capex you guys are likely to put out just going forward.
Jeff Lucas
Sure. So the lion share of the -- go ahead, Ben. You take it, please.
Ben Gagnon
Yes, I think I can just try and give like some high level figures here just to help you with your modeling estimates. As a rule of thumb, you should look at the miners as costing approximately 80% to 90% of the total capex build for a new mine. This is assuming that of course, you're buying newer miners with higher efficiency and you're not trying to look for a used miner deal at significantly lower prices. So ballpark, about 80% to 90% for the miners.
Easy way to model that would be splitting that into maybe like a 50/50 component. Different miners and different manufacturers have a different cost structure and payment schedule, but one thing that's consistent with all of the miner purchase agreements is that there is usually a cost to secure the purchase in the form of a non-refundable deposit. And then there's a cost to pay out like a settlement balance payment prior to pick up.
So those costs will need to be incurred in advance. And then when you look at the infrastructure that makes up that 10% to 20% remaining, I would look at that as almost like a 50/50 as well just for rule of thumb. You can look at about 50% of the cost would be associated with the substation and longer lead time items, and then about 50% of that cost is going to be associated with labor and other component pieces that are not long lead time items throughout the course of the build.
And in that way you can kind of use those percentages as a rough rule of thumb for looking at how capex spreads over time, but each project and each purchase is going to have its own unique variable. So it's kind of a hard question to answer specifically.
Jeff Lucas
Yes, Ben characterized that well.
Martin Toner
That's very good. That's great color, thank you very much. What type of balance sheet levels would you -- is Bitfarm comfortable maintaining?
Jeff Lucas
Balance sheet levels of what?
Martin Toner
Just amount of liquidity on the balance sheet. As you guys are outlaying some of this capex like the more bullish -- I'm assuming the more bullish you are on the company, the more undervalued you think the company is, the more you'd be willing to reduce levels of liquidity and utilize ATM less. I'm just wondering what type of -- are there certain levels of liquidity that you want to maintain?
Jeff Lucas
Sure. So it's actually variable and the reason it's variable here is largely because as we are pursuing further opportunities, including in the United States here, we want to make sure that we have the liquidity and the funds on hand to really take advantage of those as the opportunity arises here. So that results in us leaving a little more cash on the balance sheet than we otherwise normally would here. But I think for us to have a base level liquidity of I'd say roughly $30 million to $50 million that we're pretty much comfortable with. $30 million or $50 million by the way, just to be clear.
But as we look at the larger picture here, we do also think about, of course, we have our Bitcoin HODL there, not that we really want to use that here for purposes of obviously liquidity, but it's there. And that gives us a little more of a cushion to be a little more caution -- conscientious, I should say, in terms of how we are deploying our capital and making sure that we maximize the return of our assets, including our liquidity assets.
Martin Toner
That's great. Thanks, Jeff. Last question from me, it sounds like the Synthetic HODL is working. Interest in an ability to increase the size of it.
Jeff Lucas
Well, what dictates the size of, it's really two things. One is our very rigorous governance process we have around what goes on. We have a risk committee comprised of the top five managers who are very careful, including Ben by the way. Who are very careful in terms of we view our next steps here. Secondly, any strategy that we do regarding any derivative activity, we have to present to the Board for their approval to make sure they're comfortable with it as well. So we have some pretty good controls in place here.
That is indeed a guardrail over how extensive we get involved in the Synthetic HODL. But most importantly from your question, Martin, what drives how much we do with that Synthetic HODL here really is based on what are our capex and cash needs going forward here, because we really want to use our cash flow from operations to fund our capex. It is dramatically cheaper from the cost of capital standpoint, of course, than going to the street.
So I can't give you a direct answer there, but I can give you a sense of those are the two things that sort of govern how we really look at our Synthetic HODL. We anticipate, by the way, increasing our Synthetic HODL going forward here because we anticipate increasing our cash flow from operations here. And as we're working to address our extensive cap needs that we've identified for this year here, we're going to be in a better position to really take advantage of both the Synthetic HODL and our traditional HODL as well.
Martin Toner
That's great. Thanks for all the candor. That's all for me.
Operator
We have reached the end of the question-and-answer session, and I will now turn the call over to management for closing remarks.
Jeff Lucas
We want to thank all of you for joining us on the call today. It's been an interesting and we are in a very exciting time for the company going forward here. I think we're very excited. We appreciate and value greatly Nico and you're stepping in here at an important time for us. Your guidance, your experience, the history you have both in mining and with the company here, has been crucial in guiding us going forward here and we encourage everybody stay touch. We are really doing an extraordinary growth mode for us, as you know here.
Just to repeat some of the key comments here, about a 223% increase in our exahash for the year. Extraordinary not only for us, but for the industry overall. And a dramatic 45% improvement, almost, in our efficiency here. So we are very excited about what's ahead of us and really keeping an eye on the long ball here. Thank you all for joining us.
Ben Gagnon
Thank you.
Operator
This concludes today's conference, and you may disconnect your lines at this time. Thank you for your participation.
- Read more current BITF analysis and news
- View all earnings call transcripts