Earnings Call Transcript - CSIQ Q3 2020

Operator, Operator

Ladies and gentlemen, thank you for standing by. Welcome to Canadian Solar's Third Quarter 2020 Earnings Conference Call. My name is Rachel, and I will be your operator for today. At this time, all participants are in a listen-only mode. Later, we will conduct a question-and-answer session. As a reminder, this conference is being recorded for replay purposes. I would now like to turn the call over to Isabel Zhang, IR Manager at Canadian Solar. Please go ahead.

Isabel Zhang, IR Manager

Thank you, operator. And welcome everyone to Canadian Solar's third quarter 2020 conference call. Please note that we have provided slides to accompany today’s conference call, which are available on Canadian Solar's investor relations website. Joining us today are Dr. Shawn Qu, Chairman and CEO; Dr. Huifeng Chang, Senior VP and CFO; Ismael Guerrero, Corporate VP and President of Canadian Solar’s wholly owned Energy Business; and Yan Zhuang, President of Canadian Solar’s majority owned subsidiary, CSI Solar Coal Limited. All company executives will participate in the Q&A session after management’s formal remarks. On this call, Shawn will go through an overview of Canadian Solar's strategy and provide an update on the listing of CSI Solar. Yan and Ismael will respectively review the highlights of the Module and System Solutions or MSS and Energy businesses, followed by Huifeng who will go through the financial results. Shawn will conclude the prepared remarks with the business outlook, after which we will have time for questions. Before we begin, may I 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 the protection of 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 the 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 a non-GAAP basis. By disclosing certain non-GAAP information, management intends to provide investors with additional information to permit 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. At this time, I would like to turn over the call to Canadian Solar's Chairman and CEO, Dr. Shawn Qu. Shawn, please go ahead.

Shawn Qu, Chairman and CEO

Thank you, Isabel. And hello, everyone. Welcome and thanks for joining us today. I am pleased to report a strong set of third-quarter results as we achieved 3.2 gigawatts in module shipments, attained total revenue of US$914 million, and recorded a gross margin of 19.5%, which was all ahead of our expectations. Turning to slide three, let me give you a quick update on the progress of the planned China listing of our MSS subsidiary. We successfully completed the pre-IPO equity raising in Q3. This was all scheduled and it has brought us one step closer to qualifying for the IPO in China from a regulatory standpoint. The transaction also brought in important strategic partners and the capital to immediately expand our capacity with the latest solar technology. In terms of timeline, we remain well within our planned schedule, targeting the IPO for next year. We feel good about where we are in the process and believe we are on track to submit the official IPO application by the second quarter of next year. Now moving to slide four. While the solar market has been volatile over the past year as a result of COVID-19, I want to congratulate the Canadian Solar team for its focus, execution, and results. We continue to achieve long-term sustainable growth and returns for shareholders. Our long-term strategy is threefold. First, on the MSS side, we are focused on expanding capacity and increasing the level of vertical integration. This will allow us to gain global market share, enhance pricing power, better control costs, and improve profitability. Yan will expand on this point later on. Second, we're expanding our market presence and have identified opportunities in localized large-scale project investment vehicles. These vehicles will hold grid-connected solar energy storage and other clean energy projects developed by our energy business, leveraging the experience we gain through the publicly listed Canadian Solar Infrastructure Fund in Japan. We target to launch similar vehicles in Latin America and Europe within the next 12 to 24 months. Ismael will expand on this point. Third, we continue to execute on strategic growth areas. An example of this is the solar PV plus storage market segment where we are reaching key milestones, including recently signing our first large-scale energy storage system supply and service agreements, the 300 megawatt-hour milestone project in California. And there are more to come. On the development side, we also have a large backlog of 1.2 gigawatt hours and an additional pipeline of 4.8 gigawatt hours in storage projects under development. We expect energy storage to start contributing to both revenue and profit over the next few quarters and become a material earnings driver going forward. Now, let me turn over the call to Yan. Yan, please go ahead.

Yan Zhuang, President of CSI Solar

Thank you, Shawn. On the MSS business, Q3 shipments were up 33% year-over-year to 3.2 gigawatts, above guidance. Revenues and gross margin of $695 million and 18.5%, respectively, were also better than our expectations, as demand came back strong and costs did not increase as fast as we anticipated. Shawn briefly talked about the strategic direction of the MSS business. Turning to slide five, please. We're making a deliberate strategic shift as we enter a new era of growth in solar energy driven by grid parity. Historically, we took a capitalized approach that gave us the flexibility to quickly respond to unexpected policy and market changes. Today, as global decarbonization efforts intensify, and solar energy continues to improve its competitiveness, we expect more stable and sustainable demand growth. This is why we're shifting our strategy towards growing market share through vertical integration to better control our manufacturing costs. Specifically, we're planning to nearly double our manufacturing capacity across the supply chain and increase our ingot, wafer, cell, and module capacities to 11, 18, and 23 gigawatts, respectively, by the middle of 2021. All new capacity will be based on the latest technology and will produce our latest 500 and 600 watt plus modules, which we introduced during the past Q3. The new capacity will contribute to our 2021 gross margin as soon as 2022 next year and will support our 18 to 20 gigawatts shipping guidance for 2021. This is a long-term growth plan that will be supported by the China listing of CSI Solar, which includes our MSS business and the China Energy business. We believe that extending access to China's deep and liquid capital markets will continue to reduce our cost of financing and boost our long-term growth opportunities. Despite the relatively strong performance in Q3, we are facing near-term challenges driven by a combination of factors. Turning to slide six please, to give you a sense of the magnitude. The cost of polysilicon increased by around 70% between June and September this year. And while it has come down somewhat, it remains around 50% higher than where it was in June. Glass and EVA prices have also almost doubled since June. Logistic costs, mostly shipping, also doubled over the past quarter following another round of COVID lockdowns. Moreover, between June and November, the U.S. dollar depreciated approximately 9% relative to the RMB. As a standalone, any of these factors would have put pressure on our gross margins. But when you combine everything together, the situation is quite unprecedented. The impact of COVID led to a weak market earlier this year, which was then followed by a massive surge in demand. This created an acute, but temporary shortage of all basic materials, including aluminum, copper, glass, etcetera. So we are still talking about the impact of COVID. However, I think we're now at the point of maximum pain, or very close to it. And things should start improving from here. Please turn to slide seven. First, we're already seeing capacity expansion in the production of shortage materials. For example, some glass manufacturers for autos or building materials are switching some of their capacity to produce solar glass to take advantage of the exceptional margins. The price of polysilicon is also already coming down as previously damaged production sites have been repaired and capacity is gradually coming back online. Second, we're increasing our market presence in China, where relative prices are attractive for the first time in a few years. This is amplified by the fact that margins are not affected by foreign exchange and ocean shipping costs. We're seeing very strong demand in China after two transition years towards grid parity and serious policy efforts to decarbonize the economy. In terms of magnitude, we think China could account for approximately 25% of our shipments next year, up from approximately 10% in an average year. Third, solar power Q4 sales include very low ASP contracts that we’ll find in Q2, when market demand was severely suppressed. We're now working closely with our customers to educate them about the supply-side situation and will continue to raise prices. Fourth, our investment in new capacity, especially in the mid and upstream manufacturing processes, will help us reduce costs and will contribute positively to gross margins as early as Q2 next year. Fifth, visibility into our energy storage pipeline both in terms of system solutions and storage project development gives us confidence that energy storage will also contribute to next year's earnings. Finally, we are also seeing fast-growing demand for self-consumption of solar energy in the higher-margin, rooftop, or distributed generation segment. This is a market where we have a very strong market positioning. In fact, over the past few quarters, more than half of our shipments were sold to the DG segment, of which around a third was in the residential market. We expect significant growth and sustainable profitability in this market as we deliver integrated solar plus storage solutions by offering a differentiated and innovative customer value proposition with strong marketing support. This is one of our important strategic directions going forward. Overall, we remain very positive about our medium to long-term outlook based on more sustainable global demand, capacity expansion, increased vertical integration, and ongoing improvement in raw material supply and growth in the DG market. While we are tackling the short-term challenges, the underlying fundamentals of the industry remain very strong, as solar becomes fully market-driven and as we come closer to the bottom of the structural cost curve. We expect margins to stabilize at a healthier level. This will be driven by technology, scale, brand, and global market position—all of which are areas where we have a competitive advantage. Before turning over the call to Ismael, let me touch on China Energy as this is part of CSI Solar. This quarter, we sold two large operating projects totaling 200 megawatts in China. This was a key driver of the revenue and profit in the energy business. After this sale, we still have around 260 megawatts in our China portfolio, of which we expect to sell around half over the next year or so while maintaining ownership of the other half. The Chinese project development environment was quite tough over the past few years due to delays in subsidy payments, but with grid parity, this problem has been removed. So we see a lot of growth potential from a market-driven business. With that, let me pass it on to Ismael. Ismael, please go ahead.

Ismael Guerrero, Corporate VP and President of Energy Business

Thank you, Yan. I'm pleased to join today's call. On the Energy business looking at slide eight, we made significant progress over the last quarter towards overcoming the challenges that arose from COVID-19. The availability of financing, including the availability of tax equity in the U.S. was a key challenge. We have now secured most of the necessary financing in order to execute on our projects. In the U.S., we started construction on two large projects in Texas, totaling over 500 megawatts. In Mexico, we closed non-recourse project financing and started construction on a 126 megawatts project. We currently have two projects under construction in Mexico, and we are working hard to complete the interconnection on one of these projects as we speak. In terms of business development, we secured 862 megawatts in PPAs in Brazil a few days ago through a private auction with a large local utility and bilateral corporate agreement with one of the largest financial institutions in Latin America. This is in addition to the 274 megawatts we found earlier in Q2. We were one of the first companies to enter the Brazilian market both as a project developer and as a module supplier and systems integrator, and we see significant potential as we continue to expand our leadership position in the key Brazilian market. In Japan, we announced our success in securing 22 megawatts in the fixed hidden tariff auction. Japan is another high potential market where we continue to have a position of leadership. While the country is transitioning from a subsidized hidden tariff market to an options market, the recent changes in the government regarding Japan's carbon neutrality pledges, review of land use planning, and connection infrastructure initiatives, we believe this will make Japan an even more robust market over the long run. Meanwhile, we still have a large 290 megawatts backlog, plus an additional 80 megawatts of projects in operation. Combined, these projects have a weighted average feed-in tariff of approximately 29 U.S. cents per kilowatt hour, which is much higher than the global average PPA. We will continue to grow our Japanese pipeline and seek additional long-term growth opportunities in other East Asian markets. In Europe, our pipeline has grown strongly and we are starting to see success in new countries where we started investing around a year ago. In Italy, despite some COVID-related delays, our projects are progressing with the permitting. In terms of Solar PV plus energy storage and standalone storage, we are solidifying our leadership position in the U.S., while internally sharing and building our expertise across our global teams. The development of storage projects is now part of our core global pipeline, and we expect to report our projects starting construction in the following quarters. In terms of project sales, in addition to the project sold in China, which Yan mentioned, we also completed sales in Canada, Japan, and the U.S. Currently, we are on track to achieve our 2020 targets despite the tough market environment. However, we cannot rule out the possibility that certain projects may move into Q1 next year, but that is only a matter of time. Moving on to slide nine, while COVID disrupted some of our near-term plans, we continue to expect 25% annual volume growth in project sales while retaining minority ownership in projects in certain markets. We expect to enrich the Canadian Solar ecosystem and capture additional value throughout project lifecycles by capturing income from operations and maintenance, asset management services, storage integration, retrofits, and EPC while recycling most of the capital. Our goal is to increase the predictability of our revenues and cash flows as we unlock the company's value for investors and start to retain a bigger portion of the value we create. To do this, we are in the process of forming capital partnerships in the form of public or private vehicles with long-term investors. The next milestone should be in Brazil. We expect to form a Brazilian Participation Fund for Infrastructure projects or FIP-IE in 2021. We are also working on similar vehicles in Europe. The specific timing will depend on market conditions, but we are on track and look forward to updating you as we achieve milestones over the coming quarters. Now, let me turn the call over to Huifeng, who will go through the financial results in greater detail. Huifeng, please go ahead.

Huifeng Chang, Senior VP and CFO

Thank you, Ismael. Please turn to slide 10. Revenue in Q3 was $914 million, up 31% from Q2, and up 20% year-over-year. Gross margin in Q3 was 19.5%, well above our guidance of 14% to 16%. If we exclude the benefit from the U.S. anti-dumping and countervailing duty true-up in Q2, gross margin would improve by 130 basis points quarter-over-quarter. Selling expenses were flat quarter-over-quarter but up year-over-year, due to higher shipping costs. G&A expenses were down 9% year-over-year. The quarter-over-quarter increase reflects net one-off benefits in Q2 2020, such as insurance gain. Overall, our G&A expense relative to revenue has been declining over the recent quarters. The net foreign exchange loss in the third quarter was $13 million, a significant hit on our net income compared to previous quarters. This negative impact was caused by the sharp depreciation of the U.S. dollar relative to the RMB, around 5% during Q3, which is the largest quarterly change of the past 12 years. While we have a comprehensive hedging program in place, covering more than a dozen currencies, the U.S. dollar to RMB exposure is by far the largest since most of our revenue is earned in U.S. dollars, while most of our costs are incurred in RMB. We have significantly reduced our U.S. dollar to RMB exposure with higher hedging positions and expect the impact in Q4 to be much smaller, despite another four percentage points depreciation of the U.S. dollar since the beginning of October. Net income attributable to Canadian Solar in Q3 was $8.8 million or $0.15 per diluted share. This was significantly impacted by the $12.6 million withholding tax expense in China related to the special dividend distribution from the MSS business to the parent company. Excluding this one-off tax impact, net income in Q3 would have been $21.4 million higher than our Q2 net income. Moving on to the balance sheet on slide 11, we ended Q3 with an enlarged balance sheet. While we increased our total debt to $2.3 billion this quarter, mainly driven by the convertible bond issuance, our total unrestricted cash balance was $1.1 billion, or more than double our usual average of $500 million. Note that there is an element of timing here as we closed both the convertible bond and the pre-IPO fundraising for our MSS business near the end of the quarter. Since then, we have started to deploy the capital raised to support our capacity expansion plans. CapEx in the first nine months of the year was approximately $180 million. We are raising the full-year 2020 CapEx plan to approximately $500 million, which includes building capacity in both modules and storage in Q4. The higher expected CapEx is well supported by the cash raised in Q3. For 2021, we expect CapEx to be around $700 million, which is significantly higher than in previous years. The higher level is in support of our long-term growth strategy with our level of vertical integration and continue to roll out cutting-edge products. Now, I would like to spend a few minutes going through the latest corporate structure of Canadian Solar. On slide 12, following the successful close of our pre-IPO fundraising at the end of Q3, we will start reporting results for different segments next quarter. Specifically, we will change the MSS versus Energy structure to CSI Solar versus Global Energy. As Yan earlier mentioned, we will move the relatively small churn energy business from Energy to CSI Solar, which makes sense from a management perspective. In terms of energy storage, we have teams on board, the CSI Solar and the Global Energy business. SSES or System Solutions and Energy Storage delivers four integrated energy storage system solutions. This team is behind the recent storage supply and service agreement announcement with the Mustang project in the U.S. This team is part of CSI Solar and expects to deliver around 560 megawatt-hours next year, including the Mustang project. On the Energy business side, which Ismael talked about earlier, our focus is on developing solar plus storage or stand-alone storage projects. Our developers have built up a sizable backlog and a pipeline of projects over the past few quarters. This structure will allow us to benefit from the synergies of having both teams, i.e. Integrator and developer under one corporate umbrella while also capturing the synergies of attaching storage to previously solar-only contracts. Meanwhile, the accounting of the revenue and the profits from the two businesses will be separate. From a modeling standpoint, it’s also important to remember that we sold 20.4% of CSI Solar to minority investors. The additional 4.7% representing the ESOP platform will only be reflected after the completion of the IPO. We will continue to fully consolidate the CSI Solar business; the 20.4% of minority interests will be subtracted from CSI Solar's earnings and reflected as income or loss attributable to non-controlling interests. Now, let me pass it back to Shawn, who will conclude with our guidance and the business outlook. Shawn?

Shawn Qu, Chairman and CEO

Thanks, Huifeng. For the fourth quarter of 2020, we expect total module shipments to be in the range of 2.9 to 3 gigawatts. Total revenue is expected to be in the range of $980 million to $1.015 billion. Gross margin is expected to be between 8% to 10%. This means that for the full year 2020, we now expect total module shipments to be in a range of 11.2 to 11.3 gigawatts, meeting the prior guidance range. Meanwhile, we are reiterating our full-year 2021 module shipment guidance of 18 to 20 gigawatts. Our updated guidance reflects the impact of the shortage and the price increase of certain raw materials. However, we expect the environment to improve in the first half of 2021 as raw material supply expands, module ASP increases, and we start production of our upstream ingot capacity, which should reduce our manufacturing cost. Despite the near-term challenges, we remain very excited about our mid-to-long-term growth outlook given the strong underlying market fundamentals, and Canadian Solar's leadership position across manufacturing, project development, and energy storage solution factories. With that, I would now like to open the call to your questions. Operator?

Operator, Operator

Certainly, ladies and gentlemen, we will now begin the question-and-answer session. [Operator Instructions]. Your first question comes from the line of Colin Rusch of Oppenheimer. Please ask your question.

Colin Rusch, Analyst

Thanks so much, guys. So historically what we've been able to do is pass on some of these supply chain costs on your customers. Shawn, could you talk about those dynamics? How long that might take? And how effective those efforts might be as you look at the next couple of quarters?

Shawn Qu, Chairman and CEO

Well, I will let Yan expand on these first and our settlement.

Yan Zhuang, President of CSI Solar

Well, actually, the demand started to come back in August. However, the costs came back even harder. So, it takes time—a few rounds of negotiations with the customer to accept this market reality. So, we feel right now that the discussion is pretty heated already at this moment for Q4 shipments. I believe that after the primarily year-end rush is down, and once the market realizes that the cost is not coming down, they will have to face the reality. So we believe that Q1 will already show improvement in terms of market acceptance regarding the module price adjustment. We also believe that by Q1, a few cost items will likely start to improve more mildly. So we expect to see improvements in Q1, and more significant improvements in Q2.

Colin Rusch, Analyst

Great. And then looking at the energy storage business, I know you're talking about that becoming profitable. Given what we're seeing in terms of demands versus availability supply, how should we think about the margins on that business relative to historic levels for the module on MSS business?

Yan Zhuang, President of CSI Solar

I cannot give you specific numbers. However, I can tell you that for any storage system integration, it is not a manufacturing business. So the OpEx level is relatively much lower, resulting in a net profit level that is actually much better than the module business.

Colin Rusch, Analyst

Okay. That's it all for you guys. Thanks so much.

Operator, Operator

Your next question comes from the line of Brian Lee of Goldman Sachs. Please ask your question.

Brian Lee, Analyst

Hey, guys, thanks for taking the questions. Maybe the first one just on gross margins. The 8% to 10% for Q4, can you give us a sense for how it breaks out between modules and/or MSS and energy?

Shawn Qu, Chairman and CEO

The MSS gross margin is slightly lower, and the energy gross margin is slightly higher, but there is not so much difference. However, in terms of revenue, the MSS revenue accounts for about two-thirds or a little more than two-thirds of the total Q4 revenue. Therefore, the 8% to 10% reflects more of the MSS or module business' gross margin.

Brian Lee, Analyst

Okay. Thank you, Shawn. That's helpful color. And I guess I know you're saying that gross margin should improve in Q1 and then significantly improving Q2. So, I don't want to put words in your mouth. But it's really the second half of 2021 when we start to see more normalized gross margins. Would you say normalized gross margins are in the 15% to 20% range? Is that kind of the new normal, or do you think we're back to a 20% plus gross margin across the business in the second half of 2021?

Shawn Qu, Chairman and CEO

That's a very good question, Brian. It will be at least above 15% as normalized gross margin. I hope it can be more than 20%, but it is too far to tell. And we do have a plan that we believe will allow us to achieve a gross margin above 10% in Q1. Hopefully in Q2, we will target and try to reach mid-teens in Q2 and Q4. Since our new capacity, especially the ingot and wafer capacity, will start to make significant contributions, we hope that the gross margin can further improve. Of course, there are still some unknowns; we can control or influence many of the micros, but it is difficult for us to control the macro factors, such as the exchange rate between the U.S. dollar and the Chinese RMB and other currencies. All of this assumption is based on a reasonable foreign exchange between the U.S. dollar and RMB.

Brian Lee, Analyst

Okay. Fair enough. And maybe last question from me, then I'll pass it on. I know you guys are making a strategic shift to do more vertical integration; this should help the cost structure. And as you said, Shawn, you're kind of controlling the micro where you can? What's out of your control, like currency and the macro, that's the uncertainty. How do you think about the macro side of things? It seems like there's been a significant amount of module capacity addition announcements from both yourselves and your peers in Tier-1 China manufacturing suppliers heading into the next 12 to 18 months. So, does that give you pause at all, in terms of potential for supply-demand imbalance heading into next year? Just we'd love to hear your thoughts around the macro and what you’re seeing competitively regarding the capacity side. Thank you.

Shawn Qu, Chairman and CEO

Thank you, Brian. Let's talk about the micro and macro. I think from a micro point of view, I have a lot of confidence in the growth potential of the solar business. The whole world is going for renewable energy and solar. On the macro side, the solar demand in a particular year also depends on macro conditions, such as prices. For 2020/2021, the solar module price and other component prices are at a reasonable level. I believe our customers can tolerate or accept a moderate solar module price increase. However, if the module price increases too much, it will cap the total demand. Most third-party research agencies are now expecting around 160 or 170 gigawatts of capacity next year. Regarding the module capacity announcements and shipment targets, we are watching that closely as well. We are not really looking at the smaller module players; we rather pay attention to the top six or top eight. The top six and top eight have announced targets for next year in the range of 161 to 170 gigawatts. If the total market demand is in the 160 gigawatts and the total shipment targets of the top six are also in that range, there may be some competition. But I believe this competition is manageable. In terms of Canadian Solar, we have been looking at our channels and technical capabilities, and when we decided to target 20 gigawatts next year, we believe we will achieve our target. I can't really comment on other firms' targets, but I do believe we will meet ours.

Yan Zhuang, President of CSI Solar

Just want to add a little bit more. We feel that the demand from both China and the U.S. would be very strong next year. We have a sizable capacity in Southeast Asia, which will help us gain more market share in the U.S. Additionally, we have a strong movement in China to significantly enhance our market share there. Given that the pricing in China is favorable due to strong demand and the fact that the exchange rate between RMB and USD, as well as the difference in ocean shipping costs, we believe that the pricing in China will be significantly better than overseas pricing next year, helping us to gain more profit and healthier profit margins.

Brian Lee, Analyst

Alright. Thank you both. Appreciate all the color and context. Thanks.

Operator, Operator

Your next question comes from the line of Philip Shen of ROTH Capital. Please ask your question.

Philip Shen, Analyst

Hi, everyone. Thank you for taking my questions. I have a question on your capacity expansion. If you look at the ratio of wafer to your total module capacity as of 2020, it's about 40%. I think you're maintaining that mostly the same for 2021 at about 44%. So philosophically, do you think you could ever increase that wafer to module capacity ratio? And why keep that where it is, whereas sales have increased from 60% to 71% of your module capacity? Thanks.

Shawn Qu, Chairman and CEO

Hi, Philip, this is Shawn speaking. Vertical integration is what we are pursuing. We are indeed increasing the ratio of vertical integration. However, it takes time, especially with ingot capacity. We could have to buy ingots and then cut them into wafers, which doesn’t significantly contribute to our gross margin. The polysilicon stage makes the most money, followed by ingots, then wafers, and finally, modules. One key driver for next year is to increase our ingot capacity. As you can see from our chart, we're increasing the ingot and wafer capacity. By the end of next year, both ingot and wafer capacity will be around 10 gigawatts, which is significantly higher than today.

Yan Zhuang, President of CSI Solar

Yes, Philip. In terms of capacity, by the end of 2021, we will have nearly 10 gigawatts of our cell line compatible for both 210 and 182. We will also have 10 gigawatts of wafer and ingot capacity, both large wafers. For module capacity, all the new modules will be compatible with either 210 or 182. We are in good shape in terms of capacity offering. Even the existing midstream capacity of 166 is still capable of competing in the first half of next year.

Philip Shen, Analyst

Great. Thanks, Shawn. As a follow-up there, I see a lot of the industry shifting to larger wafer and cell formats. What percentage of your modules do you think will have the larger format, either 182? I believe that's your standard—versus the 210? What do you think that average is for the large format of your shipments in 2021? Do you think you can reach as high as 50% of your shipments in 2021 for the higher 182 millimeter formats? Or will it be a little bit less than that or perhaps more?

Shawn Qu, Chairman and CEO

Yes. We’re asking Yan to get into the detailed numbers. However, all of our new capacity, including new ingot and wafer capacity, as well as new solar cell and module capacity, is designed around the larger wafer and cell formats. Canadian Solar is actually a leader in adopting new formats. We were the first to introduce the 166 millimeter format and now are among the first to introduce 182 and 210 millimeter formats. By the second half of next year, and into 2022, you would expect that more than half of our product from the new formats. For 2021, it is a transitional year. So, some products will be introduced gradually by Q1, Q2, and Q3. The smaller wafer size does not mean outdated; some products designed around the smaller size may still be more suitable for certain projects.

Yan Zhuang, President of CSI Solar

In terms of capacity, by the end of 2021, we will have approximately 10 gigawatts of both ingot and wafer capacity, both large wafers. Most of the new capacity will still be in China, but we have some upgrades in Thailand and are actively exploring other markets depending on policy shifts, primarily in the U.S.

Shawn Qu, Chairman and CEO

By the way, we have more than 3 gigawatts of solar cell capacity in Thailand, which is a relatively new capacity. Looking at the total market size in the U.S., I believe that our Thailand cell capacity, combined with our Vietnam module capacity, is more than adequate to support our U.S. customers.

Operator, Operator

Your next question comes from the line of William Griffin of UBS. Please ask your question.

William Griffin, Analyst

Great. Thank you. So just wanted to ask about the capacity expansion. Historically, you guys have been committed to this, what you've called the reverse pyramid structure, which I think has been advantageous in a declining price environment. I'm just curious, why pivot away from that now if we think that the disruption in the market for raw materials is going to be temporary?

Yan Zhuang, President of CSI Solar

Shawn has mentioned that he's very confident about the sustainable growth of the industry. By analyzing the LCOE (Levelized Cost of Energy) of solar around the world, along with the strong policy shift toward decarbonization, we are quite confident about the industry's mid-to-long-term growth potential. Part of that capacity expansion is planned for next year, but more is aimed for the year after, and the years beyond. Next year may be a transition year, with very strong inherent demand at the project level and some bottlenecks in supply stemming from COVID-19. The industry is aware of this reality and is expanding capacity to meet the expected demand in future years. The shift towards market-driven dynamics and the decreasing costs of solar technology, coupled with energy storage advancements, indicates that demand is becoming less volatile and more sustainable.

Shawn Qu, Chairman and CEO

Hi, William, this is Shawn speaking. I’d like to add a few comments. Yes, our strategies in the past used to be the reverse pyramid, and we have benefited greatly from that structure. However, the operative model is now changing as the market evolves. We believe the consolidation phase we are entering will require more integrated capacity to compete effectively. We are changing our business model accordingly. I want to note that even amidst current market conditions, our manufacturing business model remains a reverse pyramid; our module capacity is still larger than our cell capacity, which in turn is larger than our wafer capacity, and wafer is larger than ingot capacity. We are thus maintaining the flexibility needed to respond to market demand changes.

William Griffin, Analyst

Yep. Very helpful. Thank you. And then just one other one on the third quarter specifically, but obviously, the margin was better than guidance and better than we were expecting. I'm just curious how the increase in polysilicon pricing in the quarter impacted you overall? I mean, do your contracts for polysilicon supply lag the actual spot pricing in the market? How does that all work?

Yan Zhuang, President of CSI Solar

The demand began to return during August after the initial lockdowns, but the price of silicon jumped due to both increased demand and a significant shortage caused by explosions in Xinjiang province at several silicon production sites. This resulted in a price increase that greatly impacted our profitability. However, we managed to mitigate some of that impact through procurement made in Q2 when prices were low. The more severe price impacts from polysilicon will likely be felt in Q4.

Shawn Qu, Chairman and CEO

Yes, William, indeed, there's a time lag from when polysilicon materials enter our ingot factory to when they become modules shipped and recognized as revenue. This can take several weeks or even a couple of months. During Q3, we strategically accumulated polysilicon stocks at low prices, which has worked in our favor. As a result, we didn't see much impact in Q3, but we expect to see that impact reflected in Q4, leading to our lowest gross margin quarter. Fortunately, we anticipate improvements in gross margin starting in Q1 and Q2 of next year, with a target to normalize by Q3.

Operator, Operator

As we are now at the top of the hour, I will turn the call back to Canadian Solar's Chairman and CEO, Dr. Qu for the closing remarks.

Shawn Qu, Chairman and CEO

Thank you for joining today's call and for your continuous support. If you have any questions or would like to set up a call, please contact our investor relations team. We hope you and your families stay safe and healthy, and enjoy the coming Thanksgiving holidays. Take care and have a nice day.

Operator, Operator

Ladies and gentlemen, this concludes our conference for today. Thank you for participating. You may now disconnect.