Keel Infrastructure Corp. (BITF) Earnings Call Transcript & Summary

Good morning, everyone. My name is Alan, and I will be your conference operator today. At this time, I would like to welcome everyone to the Bitfarms Second Quarter 2023 Financial Results Conference Call. [Operator Instructions] As a reminder, this conference is being recorded today, August 8, 2023. I would like now to turn the call over to David Barnard from LHA Investor Relations. David, you may begin your conference.

Great. Thank you, Alan. Good morning, everyone, and welcome to Bitfarms conference call for the second quarter of 2023. With me on the call today is Geoff Morphy, President and CEO; and Jeff Lucas, Chief Financial Officer. Before we begin, please note, this call is being webcast live and an accompanying the presentation. To watch along with the slides, you can log on to our website at www.bifarms.com. under the investors presentation. If you prefer to listen to the call on your smartphone, you can download the presentation from there as well. I would like to remind you that this morning, Bitfarms issued a press release announcing its second quarter 2023 financial results. Turning to Slide 2. I'll remind everyone that certain forward-looking statements will be made during the call and that future results could differ materially from those implied in these statements. The forward-looking information is based on certain assumptions and is subject to risks and uncertainties, and I invite you to consult Bitfarm's MD&A for a complete list of these. Also during the call, reference we made to supporting slides, and you can find the presentation again on our website at www.bifarms.com under the Investor Relations section. This company will also refer to certain measures not recognized under IFRS and that did not have a standardized meaning prescribed by 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 the company's second quarter 2023 MD&A for definitions of the aforementioned non-IFR measures and their reconciliation to IFR measures. Please note that all financial references are denominated in U.S. dollars unless otherwise noted. During today's call, CEO, Geoff Morphy, will review our operations for the quarter. CFO, Jeff Lucas, will follow with a detailed financial review and Geoff Morphy will return for some closing remarks after the Q&A. We have also requested investors to send questions in advance, which I will read to management after we open the call to analysts interested in the live Q&A. And now it's my pleasure to turn the call over to Geoff Morphy.

Thank you for joining us today. I'll begin by emphasizing that we have built a quality portfolio of assets and the resources to manage them very effectively. Together with ongoing investments, they provide investors with excellent exposure to rising Bitcoin prices, particularly as we approach the next Bitcoin [indiscernible]. To capitalize on this further, we are executing an aggressive and disciplined growth strategy to optimize overall returns while managing risks. Tailors of our growth strategy are responsible capital deployment, reinvestment in our fleet with the latest in minor technology, continued geographic diversification, adherence to strong financial and operating controls audited by a big 4 accounting firm since incorporation over 5 years ago and we have the most highly experienced and accomplished team supporting our global growth. In the past 24 months, we have expanded from 69 megawatts, powering 5 production facilities in Quebec to 212 megawatts, energizing 11 operating farms in 4 countries. During this time, we increased our hash rate over 275% to 5.3x as per second. We have built an incredibly sound infrastructure to support scalable and sustainable growth, and we are now leveraging our infrastructure as we pursue current and future opportunities. I'll now review our foundation. Our farms are well-established and efficient, maximizing output while minimizing costs, leading to higher and more consistent profitability with each location contributing positively. We excel in operations and continuously work to improve facilities and processes, including the adoption of cutting-edge technologies to enhance efficiency and productivity, driving effective deployment of miners and fleet enhancements. We carefully manage our capital structure. And this quarter, we further deleveraged our balance sheet by utilizing our surplus free cash flow after debt service while still investing in growth. We ended June with only $16 million in debt, enhancing our financial flexibility to expeditionally seize upon growth opportunities and deploy capital quickly and effectively. A perfect example is our accretive expansion in Paraguay, where we acquired 150 megawatts of hydro power contracts in July and recently initiated deployment plans for the first 50 megawatts, which I'll detail in a moment. We are truly a global player with a well-diversified portfolio of production assets and management personnel. These attributes mitigates risk associated with concentrated operations in a small number of facilities in a single region. Our stellar corporate development team is actively seeking new opportunities, including expanding into new geographies, adding complementary businesses and improving overall operations. With just 8 months until the next Bitcoin halving, the pipeline is growing. And with our global footprint, sound balance sheet and respective integrity, Bitfarms is well positioned to take advantage of situations meeting our criteria of stable low-cost power contracts, big paybacks and higher return on invested capital. Simply put, our portfolio of high-quality assets offers exposure to rising Bitcoin prices and Bitfarms leverages this inherent value with efficient operations, strong execution, financial discipline, global diversification and a proactive approach to identify and act on opportunities and risk management practices incorporated to safeguard against potential adverse developments. On Slide 5, I'll review some of our accomplishments for Q2 2023. We ended June 2023 with 5.3 exahash per second, up 10% from March 31, 2023, and up 47% from June 30, 2022. Current installations in Rio Cuarto and [indiscernible] continue, and our target for completion in September, increasing our exahash 19% to 6.3 exahash per second. During Q2 2023, we mined 1,223 bitcoin. With higher average bitcoin prices, Q2 2023 revenue increased 18% from Q1 2023. We delivered adjusted EBITDA of $8 million and increased our hotel to 594 Bitcoin at June 30, 2023. Slide 6 shows a summary of operating capacity and installed miners for our farms. 11 operating in 4 countries, and we now consider our 2 new locations in Paraguay as farms under development. In July, we reached 212 megawatts in operating capacity, up 28% from a year ago. Notably, 86% is powered by sustainable hydroelectricity. I will now review our operations and development plans. Turning to Slide 7. In Paraguay, we acquired 2 power purchase agreements totaling 150 megawatts of low-cost hydropower in July that will energize 2 new farms. These were very strategic acquisitions because Paraguay is highly attractive for development. Based on our experience, this country has amongst the lowest build-out costs, quickest project time lines to completion and a straightforward importation regime. Yesterday, we announced our initial deployment at our new Paso Pe farm. The Paso Pe Farm will be only about 1 kilometer away from our existing Villarrica firm. In addition to plans for 30 megawatts of air cooled facilities, we purchased MicroBT, hydro-cooled miners and related containers entirely with vendor credits or 20 megawatts of deployment of this latest mining technology. We expect the new farm to be fully commissioned at 50 megawatts in Q1 2024. Turning to Slide 8. In Rio Cuarto, Argentina, we increased production to 29 megawatts as we imported and installed approximately 5,100 new M30s Whatsminer miners. This added approximately 510 petahash per second to the facility and brought its total hash rate to approximately 700 petahash per second. We also qualified and strategically became a self-importer of miners. Doing so streamlines and lowers the cost of importing miners into the country. And as of today, 4,680 additional miners have arrived and are being processed by Argentine customs. An additional 2,797 miners are in transit, and all miners are expected to be installed and running in Q3 2023, which will bring us up to 50 megawatts at this farm. We can provide more details on this equipment, if so desired. In Canada, we closed the purchase of the Baie-Comeau acquisition and initiated production in early July. The acquisition not only brought new production capacity, but also provide us with the opportunity to optimize our fleet by redeploying miners from Magog. This was undertaken to free up suitable rack space in Magog for higher performance miners. The Baie-Comeau farm is presently operating with 1,300 miners at 5 megawatts. And with the remaining miners from Magog, we should reach 11 megawatts in Q3 2023 as planned. This, in conjunction with the Rio Cuarto build-out that I just mentioned gives us confidence we can achieve our 6.3 exahash per second Q3 2023 target. Regarding Magog, in July, we leveraged our assets to respond quickly to a lightning strike that took out our primary transformer. Our proactive risk mitigation strategy handles an unforeseen incidents and minimizes their impact on production and operations. By combining redundancy, geographic diversification, in-house capabilities and spare equipment, we were able to respond swiftly and effectively to the outage, ensuring minimal disruption to our overall business. Favorably, the facility is hashing at full capacity and no miners were damaged as part of this slighting strike. In Washington State, we upgraded intake and exhaust systems, greatly improving efficiencies. The intake is now equipped with a 2-stage filter, which includes an evaporative cooling component, which saves energy and reduces service requirements. The exhaust system now includes automated vans, reducing power consumption by as much as 90%. Results have been impressive. For example, deterioration of hash rate on a 100-plus degree day has been limited to only 2% to 3% versus approximately 30% previously. Considering this, we are now evaluating these enhancements for use on our other farms. We continue to enhance our MGMT proprietary software. This software remains one of the longest running and robust systems in the industry. New capabilities include the precise tracking of power consumption and operating performance per minor per location. This enables greater detail on minor performance, the optimization and reconciliation of electricity consumption and adds the predictive capability to power forecasting for all micro VT miners, which comprise about 90% of our fleet. In summary, as we execute against our fleet expansion and upgrade plans, we are projecting 20% sequential growth in our hash rate in Q3 2023 and with Paso Pe expected to come online in Q1 2024. We are expecting to achieve 7exahash in Q1 2024. Those are our near-term goals as we continue to evaluate other diverse capital-efficient development opportunities. Please turn to Slide 9. With that, I will now hand over the call over to Jeff Lucas for the financial review.

Thank you, Geoff. I'll begin by highlighting some key elements of our financial strategy and position. We have efficient operations and stable and predictable energy rates that with over 85% hydro are not subject to the energy cost variability associated with fossil fuels. We have a laser focus on rapid payback of capital. We enjoy low capital requirements necessary to meet our near-term growth plan, including using our existing equipment credits to reduce the capital expenditure funding needs, and we enjoy the strongest balance sheet in the history of our company that gives us the flexibility and enables us to utilize our operational expertise to take advantage of attractive growth opportunities and positions us well for the unpredictable economics that they’re having. I will now review our mining economics, our performance and our balance sheet. Turning to Slide 10. In the second quarter of 2023, we mined 1,223 Bitcoin compared to 1,297 in the first quarter of '23 and 1,257 in the second quarter of '22. The differences reflect increases in average total network difficulty by 24% sequentially and 67% year-over-year, offset by our hash rate, which was 10% higher sequentially and 52% higher year-over-year. Our second quarter revenue was $35 million, comprised of $34 million from our mining activities. This compares to $29 million from mining in the first quarter of '23 and reflects a 24% increase in the average bitcoin price quarter-over-quarter, partially offset by 6% of fewer Bitcoin mine during the quarter. Focusing on our mining economics, we turn to Slide 11 here. In the second quarter of '23, Bitfarms direct cost of production per bitcoin was under $15,700. That's up from $12,500 per bitcoin in the first quarter of '23. The change reflects the aforementioned increase in network difficulty and approximately 7% higher energy costs quarter-over-quarter. While we benefit from a low-cost stable hydro power across our productive capacity in Quebec for the first time in several years, we had a rate increase that resulted in 6% higher energy costs, with Quebec representing about 3/4 of our total Q2 '23 production, the impact was significant. That said, our increasing geographic diversification paid off as the impact of rate adjustments in one jurisdiction has moderated. I'll add one more caveat for those building financial model. Our direct cost of production since February 2022 includes a 15% accrual for value-added taxes on Canadian energy costs, reflecting proposed legislation, but which has not yet been determined or legislated. Excluding that accrual in the second quarter of '23, our direct cost of production would have been about $14,000 for Bitcoin, $1,700 less than our reported direct cost of production. Second quarter gross mining profit was $14 million or 42% of revenue compared to $12 million or 42% of revenue in the first quarter. The total cash cost of production per BTC was just under $21,800 in the second quarter of '23, up from $17,600 in Q1 '23. The largest contributor to the increase was higher network difficulty, which results in higher energy cost per Bitcoin. General and Administrative or G&A expenses were also higher over the prior quarter as a result of costs associated with moving miners among our farms to optimize efficiency, higher professional service fees related to corporate development and prospective due diligence work and business taxes in Argentina that were incurred and paid during the quarter. The quarter-over-quarter comparison also reflected benefits in the prior quarter from a onetime insurance refund and a reversal of an accrual associated with the dismissal of noise penalties at the former de la Pointe facility in Sherbrooke. Going forward, with halving in mind, we will continue to focus on reducing our G&A cost structure and have already identified savings in insurance in other discretionary areas. Please now turn to Slide 12. For the second quarter, our operating loss was $25 million, including noncash depreciation expense of $21 million and an impairment on short-term prepaid deposits and PP&E of $10 million. This compared to an operating loss of $15 million in the first quarter of '23, including a $3 million reversal of revaluation loss on digital assets, a $2 million loss on disposition of PP&E and a $1 million realized gain on disposition of digital assets. Our net loss for the second quarter was $25 million or $0.10 per basic and fully diluted share compared to a net loss for the first quarter of '23 of $2 million or $0.01 per basic and fully diluted share. Higher bitcoin prices contributed to improved profitability, with adjusted EBITDA increasing from $7 million in the first quarter of '23 to $8 million in the second quarter of '23. Profitability in the quarter was $6,200 per bitcoin versus $900 per bitcoin in the first quarter of '23. Turning to Slide 13. At June 30, we had cash of $31 million and 549 BTC valued at $17 million for total liquidity of $48 million. This compares to $42 million of liquidity at March 31, '23. In summary, of the 1,223 bitcoin reminder in the second quarter, we sold 1,109 to generate $31 million of proceeds to fund our operating and debt service requirements and deposited 114 BTC in treasury with the June month end value of about $4 million. In July, we deposited another 45 bitcoin treasury, increasing our bitcoin in custody as of July 31, '23 to 594 Bitcoin. This represents a total value of approximately $17 million based on the Bitcoin price that day of just over $29,200. In the second quarter of '23, we also raised $22 million in net proceeds from our ATM program. For the third quarter of '23 through August 7, we have raised additional net proceeds of $26 million. These monies that we raised under our ATM are specifically earmarked for the growth initiatives, which Geoff spoke about earlier. We continue to use cash generation operations to deleverage our balance sheet. Total embeddedness was reduced to $16 million at June 30 and to under $14 million at July 31. As we've noted in previous earnings calls, our debt is scheduled to be fully repaid by the end of February 2024, well in advance of the halving. With that, I will now turn the call back over to Geoff.

Thank you, Jeff. Before I open the call for questions, I would like to mention some upcoming events, especially our Analyst Day on September 14, which will begin at 8:00 a.m. at the Convene in New York City. In addition, we will be at the Canaccord 43rd Annual Growth Conference in Boston on August 9 and 10, the third Annual Needham Virtual crypto conference on September 7, the H.C. Wainwright Conference in New York, September 11, 12 and 13, and we will also be presenting at several industry events in Europe this fall. In summary, Bitfarms remains focused on accretive and diversified growth while further optimizing our facilities, fleet investments and infrastructure, including the adoption of innovative technologies and practices to enhance efficiency. Our core strengths include a competitive low-cost structure, stable and surplus sources of energy with attractive pricing, proprietary mining and facility management software, a vertically integrated electrical subsidiary and an exceptional management team. Our growth in 2023 in advance of the halving will come with minimal capital outlay given our full use of our MicroBT hardware credits. We expect to meet our near-term 6.3x exahash per second target before the end of September and 7 exahash per second in Q1, 2024. The first 50 megawatts of our 150-megawatt expansion in Paraguay which is underway and opportunities that meet our criteria for growth both before and after the halving are abundant. Operator, we can now open the call for questions. Please go ahead.

And just before you do it, I just got a note from David. I understand that we have a few questions from online investors. Let's handle those questions first, if you would, and then we'll go to the analysts.

Sure. Thanks, Geoff. I'll read them -- there's 2 questions, I'll read them both at the same time and then you can answer. One of the questions is how greatly do you think the Bitcoin halving will impact financial results in the coming fiscal year in 2024? And then another question, totally different. Does Bitfarms run any third-party firmware.

Yes, interest.

Go ahead.

Okay. Let's start with the halving. The halving we've seen it before, they happen in every 4 years. It's going to be a big event. And nobody is fooling anybody here is that we have to get -- miners that want to survive the halving need to be well prepared. We have a cost structure that is as good as anybody in the business. And as we've said in the past, we are planning to get our debts completely repaid so that we don't have P&I payments. We plan to get all our construction finished so that we don't have construction commitments. We plan to get all our miners bought and installed so that we don't have CapEx commitments so that when you go into that halving and frankly, right now with bitcoin prices just below $30,000, our cost to mine a bitcoin on a cash basis is $15,700. If we get the accrual back from the Canadian government, then it will be $14,000. But it just goes to show that, that margin will quickly disappear. I can -- we all see the analysts talking about margins and how other people's costs are higher, some are lower, but some are much higher. They are going to be extremely stressed. And typically, as we've seen in the past, the adjustment to the halving takes 5 or 6 months. I think my gut is telling me this time that it might be a little quicker because there's more adoption and more knowledge in the industry. And that 4, 5, 6 months later, bitcoin prices go up and then things get really exciting. But in the meantime, you have to be battened down and ready for a storm, frankly. And we for example, it's because of that discipline and our orientation that when bitcoin prices fell below $16,000 in the fourth quarter of last year, we don't want a very few miners with a positive adjusted EBITDA. And that goes to show what happens. I think if you really want to test what miners were go back to the fourth quarter last year and see their results. We're well positioned. And I think sort of as we go into the next summer, when the halving is going to take place without debt, we're going to even be better positioned. And with the Paraguay expansion, those are going to be low cost, we are going to have more cash flow. And the other thing you have to remember, when things get adverse like they do, miners with inefficient operations go offline. So that's when the network hash rate goes down, our market share goes up, our block rewards go up, and there's a compensating factor there. We think we're well positioned. So that's the first question. Okay. Firmware. Yes, because we are so operationally focused, we have people within our operations that look at these type of things and test them all the time. So it's getting better control of your miners and squeezing more output from them. Of course, we all want to do that. Sometimes it's firmware and now there's some hardware fixes with cards as well. We test a number of them because there's advantages to that. Sometimes the OEM firmware is fine. And the other thing that we need to keep on track, which is a little bit different from some of the others. About 90% of our fleet is MicroBT versus Bitmain. So some of these firmwares are -- firmware changes are more tailored for Bitmain, some work with micro BT, but yes, we're always looking at these as trying to get an advantage. So -- and it's not necessarily uniformly applied across the fleet. We do not want to fall into a bug or something like that. So we test them on a very limited basis and roll them out a little more when appropriate. So hopefully, that addresses those 2 questions. So operator, let's go to the analysts, please.

[Operator Instructions] Our first question comes from Josh Siegler from Cantor Fitzgerald.

I guess the start should you help us understanding of how you're thinking about future growth, specifically, which geography is really going to drive that growth. With the new Paraguay contracts in place, there's still opportunity in Argentina, how are you going to balance between the 2 geographies when thinking about scaling in the future?

Josh, we've got management and personnel in every single geography. It makes sense to have critical mass in every geography that we operate in, so that we can do that. And we, over the last 2, 2.5 years, have done an active technology and knowledge transfer amongst employees so that we have Argentine employees and Washington employees coming up to Quebec. Quebec employee is going to the other geographies and teaching. So those best practices that we developed in Quebec years ago, they're now embedded in our other geographies. And we now feel we've got critical mass in Argentina. We've got a good team, a solid team in Washington and, of course, in Quebec. So now the opportunity came about to really take our 10-megawatt operation in Paraguay and build it up. And frankly, we're excited about Paraguay. It was a site where it was a bit of a test add back in late '21 when we built it, it became active in January '22. At 10 megawatts, we shipped down used miners for Quebec. The experiment was a very good success. We put that facility together. It's a warehouse style building in about 3.5 months. It was our lowest cost infrastructure to build, and we subsequently put brand new miners in there and brought the production up. But our plan is to hire more personnel in Paraguay for this build and build establish critical mass. The Paso Pe facility is 1 kilometer from our Villarrica site so we can already take advantage of some of the talent there. But this is a much bigger build-out. It's 50 megawatts. It involves the build of a substation. So we're going to high voltage and knocking it down. The 50 megawatts is what's been announced. We've announced 20 megawatts using the micro BT hydro containers and those orders have been placed. I guess as I said in my script, the -- we're feeling very confident about the 6.3% growth for September because that's going to be from the miners coming into Argentina right now and from taking [indiscernible] from 4.85 megawatts right now up to 11, which is the maximum capacity of the electrical current available. It will be upgraded next year. And then with the 20 megawatts of contracts that we've done, that will take us up to $7 million. Civil work is going to start probably in early September there. Miners are going to start arriving later in the year and into January. The longest contract there is the transformer, the big main transformer there. We were actually overbuilding it with hope that we can get more expansion in the area. So instead of 50, we're building -- we're buying an 80-megawatt transformer there. And it's got the longest lead time, about 6.5, 7 months. We plan to put that order in the next number of days, get that started. It's actually going to be the last thing to arrive. So we expect to go from basically 0 megawatts to 50 megawatts literally with the flip of the switch. Hopefully, that's in the latter part of February, maybe it's early March, but certainly before the halving. And that's for 20 megawatts. And then there's 30 megawatts that we have not talked about yet, and that 30 megawatts, we hope to provide more news in the coming days and weeks. But this isn't guidance, but that could very much take us up to 7.8 exahash if things come together as we expect. Our guidance is reading now for 7 exahash in the first quarter of next year. I added a little bit extra there, but feel free to have a follow-up.

No, I appreciate that. And then I also wanted to touch on, as we're heading into the halving and mining rigs in secondary markets at least still remain at fairly depressed price levels. Are you thinking about potentially replacing some older machines and improving the efficiency across the fleet?

Josh, all the time, all the time. That's the nature of the business we're in. If you do not upgrade your fleet, frankly, you're going to die and halving is going to expedite that process. So I think we have some M31s right now. Some of them we're redeploying but other ones we're selling. But most of those are coming out. They're the least efficient of our miners, and we continue to upgrade. Case in point, these hydro miners are amongst the most efficient in the world with the highest output, they are going to very much enhance our efficiency across the fleet. And just like what I mentioned about Washington in terms of the changes to the intake and exhaust systems there and bringing the temperature down and getting more out of our existing miners. But our fleet, if you were to look from top to bottom is already modern, and we are continuing to modernize it and go with the latest models. We are not going for immersion cooling, but we're going to air cooled, which we do very, very well, particularly with our geographies and these hydrometer, we're very excited about that.

Our next question comes from Kevin Dede of HCW.

Mr. Morphy, Mr. Lucas. I guess the first thing is, maybe you peel the onion back a little bit on the PPA spur, I guess, in Paraguay. Are they both with Andy? Are there -- can you talk at all to, I guess, the costs associated with them and whether or not there are curtailment opportunities for you there?

Okay. So the first answer to the question is, yes, there's -- we bought companies that owned -- 2 separate companies that owned power purchase agreements. They are both with Andy. So the Villarrica site is with [indiscernible]. So we moved from the regional distributor to the national distributor. The cost of the power for that is $3.9. And as I've talked about before, there's some optimism and hope that with the new party and the election and the promises that were given that, that power might actually be reduced, like the cost of it. So fingers crossed there, and it could be a tangible reduction. So that -- we haven't banked on that, but that's what we're hoping for. In terms of the deployment, I think we haven't put out any big -- any of the numbers on that yet, except for the fact that the 20 megawatts of MicroBT hydro containers we used our credits. So from a cash -- there's still a cost there. But from a cash perspective, we're not out of pocket. So really, it's the substation that has a cost to it. The air cooled warehouses have a cost to them, but we do that very efficiently. And in Paraguay, it's frankly, the lowest cost in our whole fleet. Jeff Lucas, do you want to come in with sort of a little more finality with some of the costs there. But the 30 megawatts not announced yet. We're kind of stuck on being able to sort of give total project cost at this point. So I give that as a caveat.

Yes. Sure. I'm more than glad to speak to that here. So for Paso Pe, what we're looking at here, we're contemplating roughly $24 million, and that encompasses not only the infrastructure build-out here, including the substation in 80-megawatt substation versus the initial opportunity of 50 megawatts here. And secondly, certain guarantees and other payments have been made about $4 million to the power provider here. But overall, we're looking roughly again around $20 million that was happening in Paso Pe, which we're contemplating $14 million being paid for the remainder of this calendar year and the other $6 million of sale will be done in early 2024.

I know Mr. Morphy, you mentioned that importing things into Paraguay was a little bit easier than Argentina. And I'm wondering if the hydro machines already manufactured, would you expect any delay getting those onto site? Or is it really exclusively just a matter of getting that? I mean, from what I understand, you're starting with a greenfield open plot of land, and that needs to be treated and the substation and the grid connection and all of that has to come in. And I'm just wondering if you would expect or should -- would it be prudent on from our perspective to assume that MicroBT might hit hurdles in getting their new machines to your site.

We really don't anticipate any problems bringing the equipment into Paraguay. It's always been pretty straightforward. Compared to Argentina, it's always been quicker and easier to get equipment, in fact, even use equipment into Paraguay and the ability to source other equipment that goes into the infrastructure. It's always been really easier in Paraguay as long as you have the right documents. You can bring it in, you pay to that, and you get it on site. Argentina, as we've learned over the last 2.5 years has been more difficult. First, we didn't have the self-importation rights because we hadn't done business there for a couple of years and didn't have the track record to apply for that self-importation. We had to rely on brokers that were expensive. They only had small allocations. And then last summer, when the government was really under strain for its U.S. dollar reserves, they stopped direct importation for people like us and in many other industries, [indiscernible] tracks. And we were stagnant for a lot of months. And our 50-megawatt warehouse there was fully constructed in October, but we weren't able to put it into sort of bring it up to full operation until earlier this year. And that was a real setback and the government didn't provide any alternative mechanisms until we could qualify for -- to be a self-importer, which fortunately we are and things are a lot better now. But they also have a government change coming in November. There is uncertainty there. And it's one of the reasons why we decided to take advantage of these opportunities in Paraguay and really make that one of our geographic [indiscernible], the whole LatAm area is a geographic hub, but putting more emphasis and more growth into Paraguay at this time. It's also all hydro too, which we think long run is going to be a win for the industry and our company.

Just switching gears to Quebec for the moment. You spoke to a rate increase. I'm wondering if that was like universally and I apologize for not having researched this. I was wondering if that was universally applied by Hydro Quebec or if it was more for a township regulated and whether or not you would expect that to be rolled back or something that continues forever going forward?

Well, fortunately, rate increases in Quebec don't happen very often. And -- but this was applied right across the board. It applied to our tariff with a few nuances. It happens to other industrial tariffs. This was pretty much right across the board. And this is Hydro-Quebec, reassessing where they are, realizing their electricity grid will become more strained as electric vehicles and greenhouses and other industries using electricity in the future surge, and they contemplate building new hydro projects down the road. There's a lot of discussion with the new provincial government in Hydro-Quebec about that. We've been intervening in some of the hearings and some of the things happening. It was actually a great opportunity for us to once again talk about some of the benefits that we can bring to Hydro-Quebec of Bitcoin mining, taking our miners and putting them in areas where the hydro is rather than reeling power line distances, using some of the thermal benefits that are available. And I think they're finally seeing that like our ability to curtail on a moment's notice can actually be a huge benefit to them, and we're unique in that feature. So there's more development that is going to happen in Quebec, but I do not contemplate another increase, at least for the foreseeable future in Quebec. But yes, this was a universal one. And it was too bad, it impacted our results and will continue to impact our results. But compared to the level of inflation, like we've had one increase, 6%. That was basically the rate of -- less than the rate of inflation last year, and this has been several years since our last rate increase. So really, we're still below the increase in inflation, which, by and large, is good.

Last question for me, gentlemen. In the past, there was some discussion of future opportunity in Washington and congrats on improving your efficiency there, but I'm curious to how you might see potential expansion in Washington State.

Washington still is a very attractive place, and it's nice that we have talent there. It's one of the hot beds of crypto mining. So there's opportunities to expand both greenfield and through acquisitions. But the one thing that happened at least in the county we're in and the utility we're dealing with is they put up the price of power earlier this year as well. So we've gone from being one of the low-cost power facilities that we have to once again sort of being in sort of more in the upper quartile or so. And so that's taken some of the attractiveness out of Washington for the time being. But similarly, we're having discussions with that utility talking about the benefits. And looking around that whole area, at Columbia River and what was built for effectively the smelters from days gone by, makes it a very interesting area. So no, we're not giving up on that area.

Our next question comes from Bill Papanastasiou of Stifel.

Yes, so for my first question, just hoping to gain some insight on the average cost of power that was realized in Argentina in the quarter. Just for the purposes of measuring the impact to margins and the road ahead, obviously, the company shifted from drawing on power from the grid in Rio Cuarto to the very attractive $0.03 power contract. How should we look at that?

Sure. Let me speak to that, and then Geoff can add a little further color to that. So we are starting to get the benefits and we did see the benefits beginning to accrue in the second quarter from that shift that you spoke to. But we didn't get all the benefit here. So we've talked about Argentina cost being in the $0.03 range here. And we didn't experience that in the quarter, again because part of that quarter was actually [indiscernible] itself. So while we didn't get that much of a savings to give you some concrete numbers here. In the first quarter, the cost of power in Argentina for us was about $0.46. In the second quarter, the average amount in Argentina, again, was about $0.44. That includes this period of time where we had in the $0.03 range, getting it right from the power provider and then also a period of time when we're getting from the grid as well. So that kind of gives you a bit of a context. So the expectation here is that we're actually going to continue to achieve rate improvements in Argentina that's actually going to lower our rates overall. Now bear in mind, another comments keep in mind here that Argentina wasn't that large a contributor in the second quarter. As we're ramping up to 29 megawatts and then going to 50 megawatts perhaps even beyond here, we're going to have a much larger portion of the total bitcoin that actually will be mined in Argentina. So that also include our benefit.

Let me add to that, too. Like as Jeff mentioned, we're moving from the 29 megawatts up to the 50 megawatts very soon. That will optimize our facility there. We're still ramping up. So there's costs to do that in terms of the natural gas and the power cost. Once we're at 50 megawatts, we can optimize better. But almost more importantly is hard to get a little cooler in as part of the world in the Northern Hemisphere in the southern time part of the hemisphere, air warning into the summer season. And starting in October, which is where the summer starts, warmer temperatures, they consume less natural gas, the price of natural gas goes down. And that's where for the 6, 7, 8 months, we are -- we should see much lower natural gas prices, which will lead to lower electricity prices, and that's where I think when you -- when we do our next quarter results or certainly in March, when we do our full year results, you'll be able to ask the same question. And hopefully, we are talking more of the -- we've had less than $0.03. We really are averaging the $0.03 type of area, and we start seeing the real margins come out of that facility. And there's one other thing. They are putting an extra pipeline up, providing more natural gas to the area in which the power plant next door to us is, which is -- which was a supply constraint in the past that was being activated in July. So it should be in a pretty good state right now. So with more natural gas coming up and not being constrained, that should help the price too.

Our next question comes from Chase White of Compass Point Research and Trading.

So -- just so I'm clear that the 7 exahash target includes only the 20 megawatts of the 150 megawatts you just acquired in PPAs in Paraguay, right? It's so there's nothing incremental from Argentina has 50 megawatts, just an additional 20 megawatts in Paraguay. Okay. Just making sure.

Correct. So yes, if we infill the 30 megawatts on that 50 megawatts as we expect, I think you'll see us bump our guidance up to sort of that 7.8-ish area, but we're not ready to do that yet. We won't -- we generally only do guidance when we have clear and firmer sightlines to how to get there.

Got you. And then in terms of obviously, that would still just be 50 megawatts of 150 megawatts plus everything else in Argentina. I mean, what would you need to see in the market for -- to want to greenlight all of your potential expansion projects? And would you generally finance this with the Bitcoin sales and selling ATM shares? And how do you think about that?

So the one thing we need to see Chase more than anything else, is a little more stability and certainty in terms of what the Argentinian outlook is. And I say that because bear in mind right now, they're in the process of going through an election. They actually have their presence of primaries this Sunday, and they had the presidential election itself on October 22. There are 4 different candidates and not to develop too much in the politics here, but there are 4 different candidates who are remarkably close in the polls. Some are advocating financial policies that soon will be very constructive for us. I just require a little more thinking involved here. And part of, by the way, what's driving the fact that their U.S. dollar reserves are actually negative to turn about $8 billion. So they know they have -- the country has to address this. That can introduce uncertainties, similar to what we saw in the pain that we experienced, obviously, last year. So we want to make sure that we're looking at great opportunities here, we're also very cognizant of the potential impact and the benefits that they halving is going to bring to make sure that we're moving very carefully and very thoughtfully from that angle. In terms of how we're going to finance it, first of all, this is Argentina. You don't finance this debt clearly. Those opportunities just really aren't available and anything worthy of a reasonable rate to our price. But the goal here primarily as overall strategy speaks to is to use our excess cash flow from operations to fund the capital expenditure down there. And then secondly, of course, to use the ATM, but we're always tapping the ATM with a very careful eye to how we apply those funds to make sure that they're accretive and then we get attractive ROI and a short payback on them. Go ahead, Geoff.

That's right. Paraguay, as I mentioned, the longest lead time is the primary transformer. It's 7 months. That gets us really close to halving. And with our own internal guidelines is that we do not want to be in construction, we do not want to have debt, all of this during those first few months of the halving, we want to be lean. We want to survive. We've got long-term goals. So when we saw this opportunity to buy these 2 contracts, we didn't see it for necessarily immediate build-out, although we were hoping we might be able to, we saw it as long term. And so that 100 megawatts is an opportunity to expand. We are looking at developing it right now. Maybe we'll do the substation sort of over the next 7 months, maybe we can add some production or maybe we just do a substation and maybe we get ready so that we have warehouses and a site ready to go sort of that 4, 5, 6 months after the halving when bitcoin prices start to rip again when the network hash rate is lower and we have more market share, and it makes pure sense because of our capital structure and what we've built over the last number of years to then just jump rate in with both feet and fully build it out. But we practice a very disciplined decision process here. And so far, the pieces are not coming together to build it out over the next 7, 8 months, but we are examining that very closely, and we're looking at a number of other things, too. So capital allocation is important and we continue to want to be a diversified company and make sure that we put our precious capital where it will do the most good. Now, I think Bill got disconnected there from the note that I got. So maybe he's back and maybe, operator, you can put him back in. I think he has another question.

We will take another question from Bill Papanastasiou.

Sorry about that. I'm not sure why it disconnected. But I just wanted to talk about the repurchase of the M53S+ machines. These are the first hydro-cooled mining units in the fleet. I believe you alluded to the fact that we could see a higher proportion of hydro-cooled miners in the overall fleet in the long run, as you continue to improve fleet efficiencies. I just wanted to gain kind of a high-level understanding of how the company is looking at hydro-cooled mining units compared to other types of ASIC uses such as air cooling or immersion cooling. Could we see more peers also move towards hydro cool mining in the future?

Well, I can't really speak for our peers, but I can speak for us. And hydro-cooled is a new technology. It's not that we are on the bleeding edge here. There are deployments and it's -- can be more efficient than -- that you do not have to plunge your miners into a bath of dielectric fluid -- sticky oily dielectric fluid, it's more efficient. But then we're going into more plumbing related type of things where you have cool water that's going through, you're cooling the circuits. And we think it's more efficient. We think it's super seeds inversion. I guess time will tell, but that's our position on it. Our version has turned out to be expensive. And I think this offers more opportunities for expansion. Like these things push out production at 2x, 2.5x more than a sort of pretty much the S19 and M30s in our fleet. The efficiencies are tremendous. So we know enough and we think this is exciting enough that this is where we're going to go next. It will not replace air cooled. Air cooled is still something we do very, very well. But as you look at supercomputers and other things, they use these types of technologies. And as we continue to push for higher levels of efficiency, you have to look at what some of the other industries have done. So we're doing that...

I was going to say -- and by the way, I was anyone [indiscernible] interrupt Geoff here, but what gets us excited not only in the performance of business changes, but the economics behind it for our benefit as well. For those of you who just aren't familiar, particularly with the M53s plus and plus, plus, you seem to have 24 watts per terahash. They are about 280 terahash per unit. And the economics of these things is well below 1/3 less -- and for example, in XP Rather, which may be a little better performance here but are dramatically more expensive. I'm sorry, go ahead.

And where I was going to go next is because you have this hot fluid, you can also take this industry to what we think is also sort of the next frontier, and that's getting -- capturing the residual heat and putting it to good use. When you have air cooled, it's very tough to transport hot air at any distance at all. We do it now in Quebec by heating some of the warehouse spaces in adjacent industries. We look for aquafarming and various other things. But in fluid form and not dielectric fluid, you can push this into other things. And I know MicroBT is looking at future generations where they concentrate more heat into the -- that it's going. And I think the ability to provide heat to buildings and other things much more effectively and efficiently is there. So while we don't know that for sure, by buying these machines and getting more familiar with them now, we are very hopeful that there's -- that we can squeeze more efficiency out by possibly selling some of the heat in the fluid form down the road. And that would be particularly attractive in possibly Washington or more so in Quebec as we get more adjacent to industries and things like that.

And Geoff, you were talking about the habit that's coming up. And I was just hoping to get your expert knowledge and your team's expert knowledge on kind of the evolution of mining as we proceed to another event of Bitcoin rewards getting halved. We've seen a lot of bitcoin mining operators transition their business from being completely self-mining to diversifying into high performance and cloud computing. And many of these players cite that they've gained a significant amount of experience at operating data centers. But Bitfarms arguably has the most long-standing experience you guys are scaling to 12 farms by early next year. How do you see kind of the industry evolving from where it is today, given a potential trend of margin deterioration.

Bitcoin is an amazing thing. That protocol with its self-adjustment mechanisms and with adoption and everything else, we absolutely feel confident that over time, the adoption of Bitcoin with the inflation in the world with governments printing money that Bitcoin has a real place and I know over the next few years, 5, 10, the adoption is going to be tremendous. That will take care of a lot of things because the price is going to go up and that will allow the industry through the having to sort of reinvent itself, the weakest players will get -- will go bankrupt. There'll be some upstarts, but then there's going to be a handful of us that are public and others, there's a lot of private companies in the world that will do very well and they will expand. They will continue to push the barriers in terms of efficiency, as I said, using their thermal properties, the residual heat and other applications, being closer to sources -- to the electricity sources to be able to work on solar and wind and these type of things in energy management, all that's going to continue to evolve. But in terms of other opportunities in data center management, we operate data centers basically on steroids, high performance, but not like Tier 3 data centers where there's redundancy and everything else, they use a lot less electricity. They need to have 99.9% uptime. And then we see AI and high-power computing that sort of comes somewhere between -- there's going to be a huge demand for that. High-power computing, we can sort of do as but we're not set up to have redundancy in terms of diesel generators to come on when we have to curtail. So there has to be a fine place for us to play. And AI is exciting. Some of the high-power computing is exciting. They talk about phenomenal growth over the next 6 years. We do have experience, but I'm not sure if it's the right experience. And frankly, we're doing that research to understand whether there might be a role for us there, but we're not going to jump really nearly into something where we don't know much about. And it changes the fundamental aspect to your commercial operations. So right now, effectively, we have no clients. We mine Bitcoin and we sell Bitcoin to the pool and then we get them and we sell them to make the business work. If you could start getting into commercial contracts, sometimes you're hosting, sometimes you're not. These contracts tend to have 2 or 3 years. You have to be much more customer focused. None of us really have those customer-facing people right now. So we'd have to build that part of the business and get ready for churn. It's not a simple overnight decision just to jump into that type of business. And if we're going to do it, it needs the type of margins, the longevity to make sense, and we're evaluating that very closely.

Bill, one other comment to make here, actually, a shout out goes to our IT team because the economics naturally tightened with the head end coming up here, managing and getting information from your miners to run them more precisely and more efficiently is going to be key. And I think our MGMT2 programming and the developments we're making to that right now really give us an edge in that area. And we think that's going to work through our advantage certainly when they have income.

That concludes our question-and-answer session. I will now turn the call back over to Geoff Morphy, CEO of Bitfarms. Please go ahead, sir.

Thank you, Alan. I would like to reiterate 3 points about Bitfarms. One, we've maintained profitable mining operations each quarter as a result of our determination to maintain stable, low energy and operating costs; two, with little capital outlay, we expect to reach 6.3 exahash per second by the end of Q3 from our existing portfolio and reached 7 exahash per second in Q1, 2024. Three, a major expansion is underway and in Paraguay, and we are well positioned to move quickly on other major as well as minor expansion opportunities that meet our investment criteria. Thank you all for attending today's conference call. We look forward to updating you with our monthly production reports as well as our other developments at our upcoming Analyst Day on September 14 and when we announce our Q3 results in the fall. Thank you very much.

The conference has now concluded. Thank you for attending today's presentation. You may now disconnect.