Ladies and gentlemen, thank you for standing by. Welcome to Canadian Solar's Fourth Quarter and Full Year 2017 Earnings Conference Call. My name is Leslie, and I will be your operator for today. [Operator Instructions]. As a reminder, this conference is being recorded for replay purposes. I would now like to turn the call over to Mary Ma with Canadian Solar's IR Department. Please go ahead.

Thank you, operator, and welcome, everyone, to Canadian Solar's Fourth Quarter and Full Year 2017 Earnings Conference Call. Joining us today on the call are Dr. Shawn Qu, our Chairman and Chief Executive Officer; and Dr. Huifeng Chang, our Senior Vice President and Chief Financial Officer. Before we begin, I remind our listeners that management's prepared remarks today will contain forward-looking statements, which are subject to risks and uncertainties. And management may make additional forward-looking statements in response to your questions. Therefore, 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. And therefore, we refer you to a more detailed discussion of the risks and uncertainties in the company's annual report on Form 20-F filed with the Securities and Exchange Commission. In addition, any projections as to the company's future performance represent management's estimates as of today's call. Canadian Solar assumes no obligation to update these projections in the future, unless otherwise required by applicable law. At this time, I would like to turn the call over to our Chairman and CEO, Dr. Shawn Qu. Shawn, please go ahead.

Thank you, Mary. We appreciate everyone taking the time to join us today. The key takeaways from today's call are 2017 was milestone year for the PV industry and for Canadian Solar. Results for both the fourth quarter and the full year 2017 exceeded our expectations. Solar demand in Q4 was strong across a diverse mix of markets, in particular led by China, India and United States. We continue to lead the PV module technology evolution. We have completed our transition to our PIII technologies, which combines the diamond wire-saw wafering and black silicon cell processing to achieve lower cost and higher efficiency of multi-crystalline solar modules. Our high-power HDM modules based on Mono-PERC technology, meanwhile, have been taking more and more market share in premium markets such as Japan. We have shipped more than 6.8 gigawatt of solar modules in 2017, sustaining our track record of setting a new module shipment record every year and continuing to hold about 7% of market share worldwide. We continue to deliver on our total solution strategy. Our solar power projects have gained a higher and higher reputation as sought-after assets by traditional investors, including the world's top pension funds and renewable energy investors. We expect a continued positive balance sheet transformation based on transactions we have already announced in the first quarter of 2018. Canadian Solar has become one of the world's most bankable energy companies. Our brand name, high-quality solar assets and track record of excellence continue to set Canadian Solar apart and give us a major competitive advantage as we increase our return on investment and build shareholder value. We are pleased with our progress. Results for the fourth quarter and the full year reflect solid execution of our business strategy for both module and total solutions businesses. We have…

Thank you, Shawn. Net revenue for the fourth quarter of 2017 was USD 1.1 billion, up 21.5% sequentially and up 65.9% compared to the year-ago period. Q4 revenue came in well above our guidance as a result of the strong demand from China, India, Europe and the United States. Gross profit in Q4 was USD 218.6 million compared to USD 159.8 million in Q3. Gross margin in Q4 was 19.7% compared to 17.5% in Q3. The sequential increase in gross margin was primarily due to lower module blended manufacturing costs as a result of efficiency progress, cost controls and materials management. Total operating expenses were $88.4 million in Q4 compared to $102 million in Q3. Income from operations was $130.2 million in Q4 compared to $57.8 million in Q3. Operating margin was 11.7% in Q4 compared to 6.3% in Q3. Foreign exchange loss in Q4 was $9.5 million compared to foreign exchange loss of $16.5 million in Q3. We recorded a gain on a change in fair value of derivatives of $7.6 million in Q4 compared to a gain of $1.8 million in Q3. Income tax expense in Q4 was $28.9 million compared to income tax expense of $6.2 million in Q3. Net income attributable to Canadian Solar shareholders for Q4 was $61.4 million or $1.01 per diluted share compared to net income of $13.3 million or $0.22 per diluted share in Q3. Moving on to the balance sheet. At the end of Q4, our balance of cash and cash equivalents was $561.7 million compared to $614.6 million at the end of Q3. Our restricted cash balance was $628.1 million at the end of Q4 compared to $539.5 million at the end of Q3. Trade accounts receivable balance was $358.1 million at the end of Q4, down from the $457.4…

[Operator Instructions]. We have the first question from the line of Colin Rusch from Oppenheimer.

Can you talk a bit about the non-silicon cost reduction cadence to the balance of this year and into next year, especially as it relates to the CapEx funding?

Most of the non-silicon cost reduction in 2017 came from the introduction of the diamond wire-saw. By introducing the diamond wire-saw, I think we can reduce product cost by about $0.02 to $0.03 per watt. We also had more internal solar cell production, which also reduced our external purchase of solar cell. As you'll know, in the past few years, especially in 2015 and 2016, our module shipment growth, significantly faster than the speed of our -- the increase of internal solar cell and wafer capacity. And therefore, in 2015 and 2016, we had more and more external purchasing. Now this trend started to change and got reversed in 2017. So most of the non-silicon cost reduction came from these initiatives. However, the silicon price, especially the polysilicon price, increased a lot last year. But it's not only the silicon cost but also some other raw material costs. For example, the aluminum, glass, all these things, we have seen the price increase, especially in the second half of 2017. Somehow those cost increase more than offset our effort to reduce our non-silicon processing costs. So by the end of last year, we didn't see that much of a reduction on the module COGS. But fortunately, in 2017, the solar module price, the ASP also sustained it. So we ended up -- we still finished Q4 with around 17%, 18% of manufacturing gross margin.

Okay. And how are you seeing that trend through 2018 and into 2019?

Well, first of all, the polysilicon price have dropped significantly in the past 2 months. Moving into 2018, in January, the polysilicon price, especially inside China, was around $19 to $20 per kilogram. And these days, it's around $14 to $15. So we have seen 25% drop in the polysilicon price. We're also seeing some reductions on additional wafer sell price. Now on the non-silicon side, some raw materials, for example, the glass price and aluminum price, got stabilized and started to ease. So I think in 2018, we will see the -- this cost reduction along the solar value chain back on track. Q1, it looks like our ASP is more or less sustained because our -- most of our Q1 orders were signed already in Q4. So on ASP side, we don't see much price down in Q1. But the material price started to ease in Q1. However, from the numbers, you'll see that our shipment of volume in Q1 is less than Q4. So Q1, we will see some facility underutilization. Now moving to Q2, Q3 and Q4, we think the module ASP will go down, but the silicon price will also go down. Now the non-silicon processing costs, that's a good question. I think it will go down a little bit, reflecting the reduction in some non-silicon materials. But other than that, I don't see that much like major reduction on the non-silicon price.

Great. And then a final one for me. Just can you talk about the traction on AC modules and modules -- the DC optimizer solution and what you're seeing in terms of premium value or premium margins with those solutions at this point?

Well, we sell very little AC modules. So I don't have this number on top of my head. But our AC module is only a small portion of our sales. So if our customer wants AC module, we will ask one of AC module or optimizer guys to piggyback with our DC module product. But it really doesn't -- it's an insignificant part of our total module sales. So I'm not an expert in this area.

We have the next question from the line of Brian Lee from Goldman Sachs.

Maybe first on the total solutions business. Shawn, of the 1.2 gigawatts in operating plants as of the end of February, you sold the 309 to KEPCO, then there's the roughly 400 megawatt portfolio being sold to Shenzhen Energy. So first question, is that presumably targeted for Q2, close the sale to Shenzhen? And then the second question would be that would imply you're left with 500 megawatts left to monetize. Is that all expected to be in 2018?

Well, we haven't given the guidance for the time for the closing because some of this approval process is not fully in our control. We are working hard, working on it. All the parties are working on it. And while we make [indiscernible] progress, and get a clear signal on timing, we will update you. So that's on the first question. Now for 2018, other than the project already connected, we are going to COD some more projects. I think in U.S., we are in construction for some projects. As far as I remember, we have around 100 megawatt project already in construction in the U.S. And we also have project in construction in Japan. If you check our press release, we give you the numbers of the project in Japan, which expect to reach COD in 2018. I think we also gave you the numbers for the project in construction and expect to be COD-ed in 2018 in some other countries, for example, Brazil and China. If you add it up, all these numbers, you will see how much project we expect to add to our operation -- operating project list in 2018.

Okay, fair enough. And then just, I guess, on Japan, since it's such a big source of profitability, it sounds like based on your comments, we should be expecting operating plants on your list in Japan plus what is COD-ing in 2018 to be part of the monetization strategy for this year. Is that fair?

Well, for the Japan project, since we already have the -- our J-REIT, the Canadian Solar Infrastructure Fund, trading at Tokyo Stock Exchange, so you can expect we drop the project into a top fund pretty fast after COD. Now for some large projects, the project, if the project for a single megawatt is in the order of, let's say, 28, 30 megawatt, for those projects, sometimes in Japan, we also do private sales rather than drop into the fund. But you can expect that we monetize the project in Japan pretty much after COD.

Okay, great. Two last ones for me, and I'll pass it on. Just first on the capacity targets, it seems like there's a subtle change, a subtle downtick here. Is that being driven by lower utilization to start the year as you commented on? And what's your visibility into utilization for Q2 versus what it's averaging in Q1? And then secondly, on the -- you mentioned module ASP declines are expected in Q2 through Q4, Shawn. But can you quantify what percent declines you're expecting year-on-year on the module side of the business?

On the capacity side, moving into this month, this month is March, we are getting to almost 100% utilization of our China capacity. Our non-China capacity -- some of non-China capacity are still -- we still see some underutilization, as we said in the press release, due to these new trade decisions in the U.S. And also in India, there are some underutilization of our non-China factories. But in China, all the factories are back to full capacity, starting from this month. Now for the module price, we expect the module price to decline this year quarter-to-quarter. I don't want to give you the number because I don't want to guide my customer how to negotiate with me. But after each month, when we report quarters, we typically provide the numbers on the ASP for the finished quarter.

We have the next question from the line of Philip Shen from Roth Capital.

I'll start off with some basic housekeeping ones, and then move on to the bigger-picture ones, if that's okay? In...

Philip, can you speak up?

Sure, okay. How many of your module shipments in Q4 were sold to external customers and recognized in revenue?

How many of the modules were shipped to external customer?

In Q4...

Is that your question?

Yes.

Yes, I don't have the numbers on my hand. But I think majority of -- almost all of our shipments in Q4 went to our customers.

Great, okay. And then in terms of module ASP for Q4, what was the rough ASP in Q4? And then how do you expect that to trend in Q1?

Q4 ASP is stable, very much at the Q3 level. And Q1, as I said in my call -- in my speech, we don't see too much change on the Q1 external ASP because most of the orders, Q1 orders, were actually signed or almost signed in Q4. But moving into Q2, we see the ASP start to go down.

Okay, great. And then for the full year guidance, how much of external shipments -- well, how many of the megawatts that you targeted, I think about 6.9 or 7 gigs, will be recognize in revenue? And how much would be sent to projects?

I think still the majority will be recognized into revenues. We budgeted several hundred megawatt to be shipped to our internal project, which means we will probably buy our energy group, my project business, will probably buy some external modules as well.

Okay, great. And then bigger picture, last year, I think 40% of the global product was mono. This year, some forecasts suggest more than 50% could be mono. There seems to be a steady drumbeat to go to more mono. Remind us again how much of your cell capacity that you expect for 2018 will be mono. And then update us on your strategic and economic rationale for focusing on multi. I know that there is an economic equation that makes sense to stay with multi. But just if you can update us on your view there and what the economic conditions required might be to justify staying with mono, that would be very helpful.

Philip, our number seems to be different from your number. Our number seems to show that more than 70% of the 2017 volume were multi and less than 30% were mono. Now moving into 2018, there are different talks. But overall, I think that you will see more mono, higher percentage of mono in China than the global market. That's more because some of China's government programs, such as the Top Runner Program, focus on high efficiency rather than that much on the cost-benefit. So somehow, it gives mono higher shares. Now in our case, our -- like 6.6 to 7.1 gigawatt of product, which we forecasted for 2018, the majority, the vast majority of it is multi. And mono is only a small, small portion. But we think with this strategy, we will sustain our margin structure. So we think that we are comfortable in our multi technologies. And I don't see much issue, no matter what other people does. But Canadian Solar will be a leader in the multi world and will be successful just with our multi technologies. Now I also want to mention that Canadian Solar is also strong in the Mono-PERC technologies. We do have Mono-PERC product into some of the premium market, for example, the residential rooftop market in Japan. So we are placing different products into different markets.

One last one for me. That's really helpful in the detail there. In terms of China demand in 2018, last year, we saw 53 gigawatts. What's your view as to the size of the market [indiscernible] in 2018? Is there growth? Is there a risk of being a little lower? But regardless of the size, can you also talk to us about the cadence of the shipments in China by quarter? Because we're seeing some reports that the typical traditional utility-scale deadline to receive the feed-in tariff may be changed from June 30, which is what we've seen historically in the past, to maybe move forward if that makes sense at December 31. So you can't get this retroactivity of installation. So talk about the overall size, the quarterly cadence and then also maybe this potential move of the deadline from 12/31 -- or 6/30 to 12/31.

Philip, regarding the total volume expectation in China this year, my guess is as good as yours. Last year, everybody -- everyone guessed wrong. And this year, the number I think I've seen so far suggests 40, 45 or 47 gigawatt, which is a little bit lower than last year's number. I believe last year's official number is 53 gigawatt, right? But for our forecast, I think this is as good as people can do. Whether 45 or 47 gigawatts, I don't think it makes much difference. Maybe the year will end up at this range, maybe higher, maybe lower. I guess, my guess is as good as yours, as I said. But we are monitoring the situation. Maybe by May, when we do the Q1 earnings call, we can discuss more. Now your second question about the deadline. I haven't heard any talk about moving the deadline. I think for the project, which are received in the 2017 quarter, they still have -- these projects still have to be finished by June 30. I haven't heard that the deadline be extended. Now if you heard that, share with me. And I will double -- I will try to get some confirmation. But meanwhile, there will be some new auctions, for example, the Top Runner project auctions, which is happening as we speak. So those projects, the official deadline is December 31. So you should see some of those projects target to COD by December 31.

Okay, great, Shawn. I'll send you the information that we have, and look forward to speaking to you about it.

Okay, thanks.

[Operator Instructions]. We have the next question from Carter Driscoll from B. Riley FBR.

First question, given you've ended the year at, at least for 4Q, roughly about 37% solution sales, do you think for 2018 or even longer term that you get closer to a 50-50 split or maybe a 60-40 split solutions with your traditional component and module sales? I'm trying to get a sense of where trends this year and then longer term, if that kind of the build-to-sell model really you're going to dominate your project business going forward. Or just trying to get a sense of how much [indiscernible] side of the business.

Well, in our press release, we did give a indication of the 2018. I think we said in 2018, we expect about 55% of revenue from what we call the module and solution business. And the balance will be about 45% coming from the so-called energy business, which is more or less the project sales. However, those numbers can fluctuate year-by-year. We'll have higher project sale this year because we have built some asset back in 2016 and 2017. And because at that time, the trend, the business model favored business model in the U.S. is the yieldco. So if you remember, at that time, we build some assets in order to do a yieldco. And then SunEdison went out of business, and then nobody do yieldco anymore, so we start to sell the project. So this year, you'll see us sell a lot of projects. But moving into next year, we may not have that many project to sell. So based on that pattern, I would say that next year, maybe our revenue percentage from the module and solutions will be higher than this year. But again, this is just an expectation. It is just my guesstimate.

Okay, that's helpful. And then can you talk about -- assuming that you do sell the project to Shenzhen Energy sometime this year, do you get a similar level of debt reduction, remind us again, relative to what you just did with KEPCO?

Yes, we will also retire significant amount of nonrecourse debt. I think for the nonrecourse debt, the percentage is more or less the same. But this time, we also retire some mezzanine debt as we mentioned. In the next deal, we will not retire any mezzanine debt because we used the cash this time to pretty much retire all the mezzanine debt, which we have taken more than 2 years ago when we acquire Recurrent. So all those mezzanine debt will go.

Okay, all right. And then just last, is there any update you can give us on the bid to go private just in terms of the steps or where you stand on the process?

Can you repeat?

Just trying to get a sense of where the process is for the bid to go private that you introduced earlier part of this year and what the steps are, the timing. Any update you can give, where it is the process being considered by the Special Committee?

For that, I would ask you to wait for our next press release. When we come to some certain milestone, we will issue a press release. I believe a month ago, right, we issued an update press release, right, which -- about that the Special Committee selected their counsel and their financial adviser.

As there are no further questions at this time, I'd like to hand the call back to speakers for today.

Okay. Thanks, operator. And thanks, everyone, for joining the call today. Q1 and full year 2017 exceeded our expectations. And we continue to execute our total solutions business and our solar module business. And we remain focused on increasing return on investments and continue to take actions. And I would like to thank you for your continuous support. And if you have any further follow-up questions after today's call, please contact us. Thanks, and have a great day.

Thank you, sir. Ladies and gentlemen, that does conclude our conference for today. Thank you for participating. You may all disconnect.