Prepared remarks
Ladies and gentlemen, thank you for standing by. Welcome to Canadian Solar's Fourth Quarter 2025 Earnings Conference Call. My name is Melissa, and I will be your operator for today. As a reminder, this conference is being recorded for replay purposes. I would now like to turn the call over to Wina Huang, Head of Investor Relations at Canadian Solar. Please go ahead.
Thank you, operator, and welcome, everyone, to Canadian Solar's Fourth Quarter 2025 Conference Call. Please note that today's conference call is accompanied with slides, which are available on Canadian Solar's Investor Relations website within the Events and Presentations section. Joining us today are Dr. Shawn Qu, Chairman and CEO; and Colin Parkin, President of Canadian Solar and President of e-STORAGE; Ismael Arias, President of Recurrent Energy; and Xinbo Zhu, Senior VP and CFO. All company executives will participate in the Q&A session after management's formal remarks. On this call, Shawn will go over some key messages for the quarter. Colin and Ismael will review business highlights for Manufacturing and Recurrent Energy, respectively, and Xinbo will go through the financial results. Colin will conclude the prepared remarks with the business outlook, after which we will have time for questions. Before we begin, I would like to remind listeners that management's prepared remarks today as well as their answers to questions will contain forward-looking statements that are subject to risks and uncertainties. The company claims protection under the safe harbor for forward-looking statements that is contained in the Private Securities Litigation Reform Act of 1995. Actual results may differ from management's current expectations. Any projections of the company's future performance represent management's estimates as of today. Canadian Solar assumes no obligation to update these projections in the future unless otherwise required by applicable law. A more detailed discussion of risks and uncertainties can be found in the company's annual report on Form 20-F filed with the Securities and Exchange Commission. Management's prepared remarks will be presented within the requirements of SEC Regulation G regarding generally accepted accounting principles or GAAP. Some financial information presented during the call will be provided on both a GAAP and non-GAAP basis. By disclosing certain non-GAAP information, management intends to provide investors with additional information to enable further analysis of the company's performance and underlying trends. Management uses non-GAAP measures to better assess operating performance and to establish operational goals. Non-GAAP information should not be viewed by investors as a substitute for data prepared in accordance with GAAP. And now I would like to turn the call over to Canadian Solar's Chairman and CEO, Dr. Shawn Qu. Shawn, please go ahead.
Thank you, Wina, and thank you all for joining our fourth quarter earnings call. 2025 was another challenging year, marked by persistent market headwinds and a shifting regulatory landscape. Through these turbulent conditions, we demonstrated strategic resilience and operational discipline. We prioritized margin and diversified our profit drivers, particularly in energy storage. Now let's review our operating and financial results. Please turn to Slide 3. In the fourth quarter, we shipped 4.3 gigawatts of solar modules, bringing total shipments to global customers to 24.3 gigawatts for the year. In response to the prolonged solar downturn, we have pivoted away from the industry's traditional focus on shipment volumes. Instead, we are concentrating on strategic high-value markets. Notably, in the U.S. market, we continue to build on our historically strong track record. In 2025, we delivered a record 8.1 gigawatts to the United States. In energy storage, the volatile tariff environment shifted some shipment volumes into 2026. Even so, we ended the year with a record 7.8 gigawatt hours of global shipments, including 3.9 gigawatt hours delivered to the United States. Given downward adjustments in both solar modules and energy storage volumes, along with lighter project sales from Recurrent Energy, our total revenue for 2025 was $5.6 billion. Gross margin improved by 160 basis points year-over-year. This was driven by a higher mix of module shipments to high-value regions and a larger share of storage volumes delivered under third-party contracts. We maintained tight control over operating expenses and achieved operating income of $43 million for the full year. However, our volatile macro environment increased FX losses and interest costs grew as we increased debt to support our IPP build-out. As a result, we recorded a net loss attributable to Canadian Solar of $104 million or $2.5 per diluted share. Canadian Solar has continuously evolved over more than two decades in the renewable industry. Global opportunities have shifted over time and we have built a strong global track record across solar manufacturing, storage manufacturing and project development. Today, we see a compelling opportunity to create value by returning to our home base in North America. In December 2025, we announced a strategic initiative to resume direct oversight of our U.S. operations by forming our new U.S. manufacturing platform, CS PowerTech. For an update on our U.S. manufacturing road map, please turn to Slide 4. Canadian Solar is spearheading the effort to reshore manufacturing to North America. In Mesquite, Texas we have successfully ramped our solar module factory to an annual production run rate exceeding 5 gigawatts, supported by 1,500 local employees. As we have previously noted, we believe the United States is best served by a resilient domestic supply chain. Consistent with this view, we are doubling our nameplate capacity to 10 gigawatt peak by the end of 2026. This expansion is expected to increase our local workforce to 1,700 employees. This will make us the largest crystalline silicon solar module manufacturer in the country. We are also pleased to report progress at our flagship solar cell factory in Jeffersonville, Indiana. In response to strong customer demand, we're expanding our initial nameplate capacity beyond the originally planned 5 gigawatt peak as we install and commission additional production lines through 2026. Phase 1 will have a nameplate capacity of 2.1 gigawatt peak and will use state-of-the-art heterojunction technology, or HJT. Trial production is scheduled to begin next month. Phase 1 will represent the only commercially operational HJT solar cell facility in the United States. Phase 2 will add 4.2 gigawatts of capacity, bringing our total U.S. solar cell nameplate capacity to 6.3 gigawatt peak. This will make us the largest crystalline silicon solar cell manufacturer in the country. We expect the trial production of Phase 2 to begin by the end of 2026. We recognize the significant value of adding Phase 2 to our solar cell facility and we believe the market does as well. This was demonstrated by our recently completed $230 million convertible bond issuance. Demand for storage in the U.S. also continues to grow strongly. The rapid build-out of data centers to support AI growth is driving increases in power demand for data center infrastructure. Our OBBBA-compliant storage solutions produced in Southeast Asia have seen very strong demand. As a result, we plan to scale resources there to increase both system and battery cell capacity throughout 2026. Hence, we will be both expanding our Southeast Asia storage manufacturing capacity and advancing Phase 2 of our U.S. solar cell factory in tandem. Given the commercial priority of these two significant investments, we are strategically delaying progress at our battery cell and BESS production facility in Shelbyville, Kentucky. Given our long-term commitment to U.S. manufacturing, Recurrent Energy will proactively rebalance its business towards monetizing in-construction and operating assets in order to optimize cash flow and manage leverage. In conclusion, I'm proud of our team for delivering strong performance this year and I look forward to continuing the momentum we are building in our U.S. manufacturing initiatives. With that, I will now turn the call over to Colin, who will provide more details on our manufacturing business. Colin, please go ahead.
Thank you, Shawn. Please turn to Slide 5. In the fourth quarter, we continued to operate against a challenging solar market backdrop. Upstream cost increases, particularly in silver, along with the costs associated with underutilization across our global solar supply chain, required careful volume management to protect margins. As a result, we delivered 4.3 gigawatts of solar modules, below our guidance. In storage, shipments were delayed into the first quarter of 2026 due to construction delays at one of our customer sites. As a result, we delivered 2 gigawatt hours, slightly below our guidance. These two volume shortfalls resulted in lower-than-expected revenue of $1.3 billion. Solar module ASPs remained at record lows throughout 2025. We have been an industry leader in pivoting away from volume growth towards value growth. By controlling shipments to less profitable markets and increasing shipments to the U.S. market, we maintained blended pricing above the industry average. In 2025, the U.S. accounted for approximately one-third of our global module shipments. Meanwhile, our storage business in 2025 faced challenges created by tariff volatility and the passage of the One Big Beautiful Bill Act. These policy uncertainties materially impacted customers' project planning. While we did not lose any opportunities, some volumes shifted into 2026 as we worked closely with customers to navigate these trials. Margins normalized in the second half of the year as we delivered more recently signed contracts. With the increase in lithium carbonate prices, we are actively managing our exposure. Today, the two most important drivers of our manufacturing margin are the mix of solar module shipments to the U.S. and the performance of our storage business. Within U.S. module shipments, domestic manufacturing is becoming increasingly important to margin as we reshore production and deliver a greater portion of our strategy through our own domestic capacity. Now let me walk you through the latest updates on our battery energy storage business. Please turn to Slide 6. We delivered 2 gigawatt hours in the fourth quarter, corresponding to $297 million in revenue after accounting for volumes delivered to our own projects. Despite the delays caused by policy changes, we ended the year with a record 7.8 gigawatt hours of energy storage shipments delivered globally, a year-over-year increase of 19%. Over the past three years, we have tripled our sales in this business. Our momentum is further reflected in our record contracting backlog of $3.6 billion as of March 13, 2026. This includes contracted long-term service agreements covering 29 gigawatt hours of projects. Today, we are the market leader in Canada in terms of contracted volume with multiple gigawatt hours under contract. We maintain a strong presence in key markets, including the United States and the United Kingdom, while scaling delivery in new markets such as Australia and Latin America. At the same time, we continue to actively engage in markets such as Japan and Mainland Europe, where we see attractive new opportunities. In terms of applications, we see a significant opportunity driven by the rapid build-out of data centers. For example, the 2.5 gigawatt hour supply agreement we recently signed with a major U.S. utility reflects the sharp increase in electricity demand driven by AI and hyperscale data center development. Battery energy storage is playing an increasingly critical role in supporting the additional power requirements of data center infrastructure. By strengthening regional grid capacity, our storage solutions help ensure reliable power for these emerging loads. We have built a dedicated team focused specifically on the requirements of data centers and related infrastructure and we are beginning to see strong results from that effort. Our focus is on delivering comprehensive power solutions, including support for data centers and long-duration applications. Our differentiation lies in providing competitive and reliable total solutions. We combine strong execution capabilities, deep experience in complex grid interconnection and commissioning, and long-term service expertise with a proven product platform and track record. This allows us to deliver the value our customers are seeking as they navigate the changing demands created by rapidly growing electricity loads. Now let me hand the call over to Ismael, who will provide an overview of Recurrent Energy, Canadian Solar's global project development business. Ismael, please go ahead.
Thank you, Colin. Please turn to Slide 7. In the fourth quarter, we sold a few small PV projects in Japan, bringing total full year 2025 sales to nearly 1 gigawatt. Revenues from operating solar and battery energy storage projects decreased sequentially due to seasonality, while power services performance remained stable. Two major project sales originally planned for the fourth quarter have now shifted into 2026. One of these projects has already been sold in the first quarter. Gross margin was further pressured by impairments to project assets within our pipeline. Moreover, without sufficient scale from project sales during the quarter, we were unable to cover operating expenses, resulting in an operating loss of $69 million. We continue to shift our business mix towards the monetization of operating and under-construction assets in order to strengthen our balance sheet and improve cash flow. As we manage the pace of construction activities, we are also optimizing our pipeline for quality, focusing on generating value from existing opportunities. Please turn to Slide 8 for an update on our pipeline. As of December 2025, we have secured interconnections for around 7 gigawatts of solar and 15 gigawatt hours of energy storage globally, excluding projects already in operation. As part of our continued effort to streamline our pipeline, we have removed projects that have been impaired this quarter. Following these adjustments, our total project pipeline now stands at 24 gigawatts of solar and 83 gigawatt hours of energy storage.
Thank you, Ismael. Please turn to Slide 9. In the fourth quarter, we delivered revenue of $1.2 billion. Revenue was below guidance due to project sales delayed into 2026 and lower-than-expected volumes in both solar and storage. Gross margin was 10.2%, impacted by project asset impairments at Recurrent Energy and inventory write-downs in our manufacturing business. Selling and distribution expenses decreased 20% sequentially, primarily due to lower shipping costs associated with reduced shipment volumes. General and administrative expenses decreased 8% sequentially, driven by continued cost control measures. Net interest expense in the fourth quarter was $39 million, up $10 million from the prior quarter. This increase was driven by a modest rise in total debt balances. Net foreign exchange loss in the fourth quarter was $15 million, driven by a weaker U.S. dollar and a stronger Chinese renminbi. Total net loss for the quarter was $131 million. Net loss attributable to Canadian Solar shareholders was $86 million or $1.66 per diluted share. Now let's turn to cash flow and the balance sheet. Please turn to Slide 10. In the fourth quarter, net cash flow used in operating activities was $65 million, driven by a change in working capital, specifically an increase in project assets, partially offset by a decrease in inventories. Inventory in the prior quarter had increased in anticipation of higher input costs. Capital expenditures for the year totaled $962 million, slightly below forecast. This was primarily due to payment timing which we expect to occur in 2026. Net cash provided by financing activities was $22 million, as we increased incremental financing to provide additional financial flexibility for the group. Of our $6.5 billion in gross debt, non-recourse debt under Recurrent Energy as of December 31, 2025, was $2.2 billion. We ended the year with a cash balance of $1.9 billion, which we will deploy prudently in line with our strategic priorities.
Thank you, Xinbo. Please turn to Slide 11. For the first quarter of 2026, we expect our Manufacturing segment to deliver solar module shipments between 2.2 to 2.4 gigawatts. For energy storage, we expect shipments between 1.7 to 1.9 gigawatt hours. We forecast total revenue for the first quarter to be between $900 million and $1.1 billion, with gross margin expected to range from 13% to 15%. Margins in the first quarter are expected to remain soft across both the Manufacturing segment and Recurrent Energy. This reflects cost increases across the solar supply chain as well as delayed project sales at Recurrent Energy. For the full year of 2026, we are issuing new guidance. We expect to deliver 6.5 to 7 gigawatts of module shipments and between 4.5 to 5.5 gigawatt hours of energy storage shipments to the U.S. market. Our solar module shipments in the U.S. are expected to be slightly lower in 2026 compared with 2025. This is due to a limited supply of solar cells qualified as non-PFE under the OBBBA during the first half of the year. The elevated cost of these cells will also affect profitability. We believe this constraint will be temporary as our own domestic solar cell production ramps during the second and third quarters. Similarly, battery energy storage shipments are expected to be weighted towards the second half of the year. Within our project development business, our focus remains on rebalancing the portfolio towards asset monetization while continuing to optimize our cost structure. Overall, 2026 will be a transition year as we accelerate our U.S. manufacturing road map to further diversify our long-term profitability drivers. With that, I would now like to open the floor for questions. Operator?
Questions and answers
Our first question comes from the line of Colin Rusch with Oppenheimer & Company.
Congratulations on being able to shift the business so aggressively here. I'm curious about the pricing environment in the U.S. What are you seeing in terms of trend lines here and long-term support for pricing that supports the aggressive capacity expansion?
Yes. Colin, this is Shawn. I assume you mean the solar pricing. Solar long-term pricing in the U.S. is stable. Our main market in the U.S. is the utility-scale project market, which typically involves large orders and a few quarters of lead time. In this market, we have seen pricing per watt go up. For example, from the beginning of the year to now, U.S. pricing on average has increased by about $0.02 to $0.03 per watt. This is mainly in response to the tight supply of OBBBA-compliant solar cells and higher material costs, especially silver. So I think this will allow us to sustain margin in the U.S. market. For storage, we also see prices stable and responding to higher lithium prices. The lithium carbonate price increased since late December and the current new pricing reflects this. That will allow us to maintain gross margin more or less for the U.S. market, Colin.
That's super helpful. And then just from an organizational perspective, as you look at a different strategy around revenue and margin, how should we be thinking about operating expenses and the baseline there and how that allows for significant operating leverage as you start to grow again on the top line?
In terms of operating expenses, we see them moving down roughly proportional to shipment volume, because a large part of operating expenses is shipping cost and overhead to support volume. If volume decreases, these operating expenses decline. In the second quarter we may see some increase in operating costs for energy storage, but at the same time we'll see reductions in solar operating costs. Moving forward, if solar prices continue to decline, the percentage of operating expenses for solar will increase. Fortunately, since January this year we have not seen absolute ASP of solar modules go down; they have actually increased. Energy storage pricing is also increasing. So as long as we continue to control operating expenses and make our operations more efficient, we'll be able to grow the bottom line. I believe you will start to see that happen beginning in Q2.
Our next question comes from the line of Philip Shen with ROTH Capital Partners.
On the project delays, I'm sorry if I missed this, but what drove the project sale delays from Q4 into '26? Can you give additional color also on the impairments that you experienced?
I'll ask Ismael to address this question, and Xinbo to add any necessary supplement. Ismael?
Thank you, Shawn. There were a couple of projects that delayed mainly due to permitting delays. One of them was already finalized and the other is under exclusivity, so it should be finalized soon, pending permitting. The impairments were driven by two main reasons. One is rapid changes in legislation in some countries; many early-stage projects no longer made sense under new regulations, so we stopped putting capital into them. The other driver was sudden increases in interconnection costs on some projects, which made them economically unviable. We decided to stop investing in those projects. Those are the two main drivers.
Can you share which countries had projects where you might be downsizing efforts, either due to legislation or interconnection cost increases?
The biggest bulk is in the U.S. because of changes in regulation. Italy was also impacted because of grid reform, and France to a lesser extent. Spain was affected a bit, but not materially.
Regarding the guidance, can you explain why the 2026 guidance is focused on the U.S.? Although Q1 had global numbers, I'm wondering if you could also share what the U.S. mix for Q1 might be, and what 2026 CapEx might be?
We provided global guidance in November last year, and this time we added U.S.-specific guidance in writing. In terms of CapEx, most of the incremental CapEx for 2026 is in the United States. There are some remaining payments for capacity elsewhere, but the bulk of CapEx is in the U.S. We also have new CapEx in Southeast Asia for energy storage and lithium battery cell production, but that capacity is mainly prepared to support U.S. demand. I'll let Xinbo provide the quantitative details.
The CapEx number is the same as we guided in the last earnings call: around $1.2 billion for 2026. There might be some uncertainty depending on tariff developments in the U.S., but the guidance remains approximately $1.2 billion.
One more, if I may. Regarding the Section 337 investigation and the USPTO review related to the First Solar TOPCon patent lawsuits, can you provide perspective on the IP situation? It looks like you're focused on heterojunction, so maybe it's not that relevant, but I wanted to touch on it.
First, we decided to use heterojunction (HJT) technology for our U.S. domestic solar cell factory years ago, before the legal challenges from First Solar. We made that decision independently because we believe we are strong in HJT technology. HJT has several advantages. It has a higher theoretical efficiency limit than TOPCon, and it supports thinner wafers which can reduce cost. HJT uses a low-temperature process—around 200 degrees Celsius—compared to TOPCon that requires high-temperature processing. Because of the lower temperature, HJT can potentially use less silver and more copper in the paste, providing significant savings in silver cost, which has become a major input cost recently. I have asked our R&D team to focus on zero-silver technology, and HJT appears to be well positioned for that transition. HJT is also more equipment-dependent and can be more automated, reducing operator-per-watt requirements and easing initial operator training in the U.S. That's why we introduced HJT. As I said earlier, we expect to see the first HJT cells from our Jeffersonville, Indiana lines by the end of this month and to begin ramp-up of Phase 1 next quarter. Regarding the legal and patent issues, I won't comment in detail on litigation, but I will say that USPTO has recently rejected many review requests, which is a change in practice. We do not believe this weakens our position in court. We are confident in our technology and our view that we have not infringed others' patents. The Section 337 process typically takes about a year or more in front of the ITC, and we'll let the process run. Meanwhile, we have flexibility: HJT is progressing in the U.S., and by the end of this year we will be ramping Phase 2. By Q1 next year, you will start to see 6.3 gigawatts of nameplate solar cell capacity in the United States. That will cover a large part of the solar cell requirement for our Mesquite module factory. We increased the Mesquite module factory nameplate to 10 gigawatts. For the remaining requirement, we may use TOPCon or PERC technology to supplement. Overall, this is a transition process, and beginning in the second half of the year you will see results from this transition. We believe we will be one of the most stable players in the global market, especially in the U.S. market.
Our next question comes from the line of Maheep Mandloi with Mizuho Securities.
On the HJT expansion and the module expansion, could you discuss the capital needs or CapEx requirements for those? Separately, on the FEOC side or compliance with FEOC rules, can you touch upon that with this new strategy—do you still need the 45X? How do you plan to address the material assistance or other rules around the foreign entity of concern language in the OBBBA?
We have largely completed the restructuring necessary to be compliant with the OBBBA by December. We reformed a new entity called CS PowerTech for U.S. manufacturing. All the subsidiaries and affiliates are 75.1% owned by Canadian Solar. Canadian Solar is headquartered in Kitchener, Ontario. We recently held our first board meeting of 2026 in Kitchener, and we'll have a second board meeting there soon in May. With this structure, we believe we are compliant with the OBBBA. In terms of CapEx, most of the 2026 CapEx will occur in the United States. There are some expansion costs at the Mesquite module factory to go from 5 gigawatts to 10 gigawatts, but most of the buildings are already in place so expansion costs are relatively efficient. For the Mesquite solar cell Phase 1 and Phase 2 together, the total CapEx will be over $1 billion, which is why much of our 2026 CapEx is focused on Jeffersonville, Indiana. Phase 1 equipment is largely in place, and we are now adding Phase 2. There are also CapEx requirements in Southeast Asia for the lithium battery energy storage factory. Those are the primary expansion areas this year.
Appreciate the clarity. A follow-up: any thoughts on targeted gross margins for U.S. solar module or battery manufacturing? And do you have orders for 2027 or 2028 at the U.S. factories yet?
We typically don't guide specific gross margins for future periods, but I can share historical context. In 2025, U.S. manufacturing and U.S. orders demonstrated gross margins of more than 20% for solar module manufacturing. In the first half of this year, margins might be tighter because of limited supply of OBBBA-compliant solar cells and because we are waiting for Jeffersonville to contribute in the second half. The tight supply can reduce U.S. volume and increase solar cell prices. However, this is transitional. Our customers have shown willingness to accept some cost increases in response to higher silver and cell costs. Moving into the second half and beyond, as we shift more into domestic solar cell production, tariff impacts should diminish. Historically, we've seen over 20% gross margin for U.S. solar manufacturing. For battery energy storage, we have targeted gross margins of around 20% or higher on global shipments. Regarding future orders for 2027 and 2028, we are continuing to secure contracts, but we do not provide detailed forward order books by year at this time.
Quick clarification on the prior guidance of 25 to 30 gigawatts of shipments for 2026—is that still intended to hold or will that be revised later?
We did not provide new global volume guidance this year. At this point, we are prioritizing profit over absolute volume due to higher solar cell costs and some supply constraints. U.S. volumes may be a bit lower than last year in the near term, largely due to supply issues before domestic cell production ramps. The conflict in the Middle East is another factor that could create turbulence, particularly affecting demand from that region. For now, we are focused on profitability and will provide updated annual global volume guidance later in the year once there is greater clarity.
Our next question comes from the line of Alan Lau with Jefferies.
Congratulations on winning major orders related to data centers for energy storage. How does that project work? Do you sell power to the utility company, which in turn supplies the data center? Do you know how big the load is for the data center?
I think you're referring to our recent press release on the 2.5 gigawatt hour agreement. Colin is the President responsible for our storage business, so I'll hand this to him to provide more detail.
Thanks, Alan. That particular project is for a major U.S. utility and is part of their strategy to build out power supply for major hyperscale investments. This is a front-of-the-meter solution: it supports regional infrastructure and strengthens grid capacity to meet increased demand from data centers. We are also seeing trends toward behind-the-meter and direct-connected data center solutions, though this particular project is front-of-the-meter. We expect to see more front-of-the-meter infrastructure projects as utilities respond to growing demand. For direct-connected or behind-the-meter solutions, we are developing total power solutions that consider load flow and response requirements specific to data centers. These requirements are stringent, so we do extensive testing, including hardware-in-the-loop testing, and configure our solutions accordingly. In some cases, the total solution may include auxiliary equipment or generation in partnership with others. Our goal is to provide comprehensive power solutions, execution capability, long-term servicing, and guarantees to meet data center needs.
Quick follow-up: for behind-the-meter solutions, do you also address grid connection requirements to stabilize load for the data center? Is that part of your package on top of powering the data center?
Yes, that's part of the total solution. It can include auxiliary equipment, power supply equipment and potentially generation equipment. We can provide parts of the solution ourselves and partner for other components. We're looking at data center opportunities with a holistic power solution approach, bringing additional value through project execution, servicing, and long-term performance guarantees to meet the specific requirements for data centers.
Ladies and gentlemen, that concludes our question-and-answer session. I'll turn the floor back to management for any final comments.
All right. Thank you for joining us today and for your continued support. If you have any questions or would like to set up a call, please contact our Investor Relations team. Take care, and have a great day.
Thank you. This concludes today's conference. You may disconnect your lines at this time. Thank you for your participation.