Ed Job - Director of Investor Relations Shawn Qu - Chairman, Chief Executive Officer and President Michael G. Potter - Chief Financial Officer, Senior Vice President and Director

Kelly A. Dougherty - Macquarie Research Satya Kumar - Crédit Suisse AG, Research Division Nitin Kumar - Nomura Securities Co. Ltd., Research Division Aaron Chew - Maxim Group LLC, Research Division Pranab Kumar Sarmah - Daiwa Securities Capital Markets Co. Ltd., Research Division Timothy Lam - Citigroup Inc, Research Division

Good day, ladies and gentlemen, and welcome to your Canadian Solar Fourth Quarter and Fiscal Year 2012 Earnings Call. My name is Bhukandar, I'll be your event manager today. [Operator Instructions] I would like to advise all parties, this conference is being recorded for replay purposes. And now I would like to hand the conference over to Ed Job, Canadian Solar's Director of Investor Relations. Please proceed, sir.

Thank you, operator. Welcome, everyone, to Canadian Solar's Fourth Quarter and Fiscal Year 2012 Earnings Conference Call. Joining us on the call today are Dr. Shawn Qu, our Chairman and Chief Executive Officer; and Michael Potter, Senior Vice President and Chief Financial Officer. Before we begin, may I remind our listeners that, in this call, management's prepared remarks 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 more detailed discussions 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 represents management's estimates as of today, March 11, 2013. Canadian Solar assumes no obligation to update these projections in the future unless otherwise required by law. At this time, I would like to turn the call over to Dr. Shawn Qu. Shawn, please go ahead.

Thank you, Ed. And thank you all for joining us on the call today. 2012 has been a difficult year for the whole solar industries, with its persistent module ASP declines, module margin pressure and escalating trade disputes. In this difficult environment, we have fared relatively well. We measure our success based on the execution of our strategy, which is to increase market share, reduce cost and differentiate our business model by expanding our total solution business while maintaining prudent financial management. For the full year 2012, we have shipped 1,543 megawatts, maintaining the track record of continuously increasing our shipment volume and market share every year. We have continued to diversify our customer base, reducing our reliance on European markets while successfully pursuing growth opportunities in Asia and North America. We are especially proud of our success in the Japanese market, where we tripled our module shipments during the year. We also significantly reduced our manufacturing cost, exiting 2012 with all-in module manufacturing cost of $0.55 per watt. This is one of the lowest in the industry and also at the low end of our target of $0.55 to $0.60 per watt which we announced in early 2012. In addition, we've maintained one of the strongest balance sheet among our peers and attracted support from leading global financial institutes, which gives us the flexibility to expand our total solution business. I want to emphasize that, ever since our IPO in 2006, Canadian Solar has adopted a growth strategy different from most of our competitors. Rather than committing capitals to the upstream polysilicon business, we have focused our investments in the mid- and downstream manufacturing steps such as sales and modules, as well as the development of our downstream total solution business. As a result, we do not have to carry the heavy baggage of underperforming polysilicon assets or large amount of unfavorable long-term polysilicon supply contracts, as some of our competitors do. This strategy provides us with a clear path back to profitability, which further serves to differentiate Canadian Solar from most other companies in the industry. In addition to establishing a cost leadership position and differentiating our business model, we also made progress on differentiating our product offering through module efficiency improvement. For example, in 2012, our proprietary ELPS sales technology achieved cell conversion efficiency in the lab of up to 21.1% for monocrystalline cells, supporting module per power of up to 280 watts on a 6-inch 60-cell module. We have been commercially producing our ELPS high-efficiency modules since Q4 2011 and are exiting 2012 with 120 megawatts of annual capacity of these high-efficiency cells. During the year, we are also proud of our successful development and launch of our next-generation ResidentialAC system, which provides customers with another cost-effective breakthrough. Canadian Solar's ResidentialAC system addresses the critical limitations of current microinverters in the market, offering 25-year reliability, power generation in high-temperature environments and the ability to work with power class modules up to 300 watts, a significant increase from the current 215-watt limit. We are excited with this product launch and have received positive feedback from our customers. Moving on to our results for the fourth quarter of 2012. At 404 megawatts, our shipments came in at the middle point of our expectations, while gross margin exceeded the high point of our guidance. Strength in demand came from Japan, U.S., and Japan is also -- and Japan is emerging as a strong new leader, driven by growth of commercial project installation. In the U.S., we are seeing many of the projects which were delayed during the year finally being completed. Our utility-scale project pipeline now stands at above 780-megawatt DC, with approximately 400 megawatt in Canada, 250 megawatt in U.S. and 100 megawatt in other regions. Importantly, we have been able to secure financing to acquire and build solar power plants from key global financial institutes, underscoring the quality and bankability of our pipeline. This is -- it is worth mentioning that we have also lined up buyers for all of our projects in Ontario, Canada. During the fourth quarter, we also expanded and strengthened our project development in Asia. In the immediate future, we see compelling opportunities in Japan, and we are pleased that our team has hit the ground running. We look forward to adding utility-scale project in Japan to our high-quality, investment-grade project pipelines in the weeks ahead. Clearly, our strategic decision early on to focus on downstream total solution business has been proven correct. This is highlighted by our momentum and success in low-risk countries such as Canada, U.S. and Japan. These initiatives will start to bear fruit in 2013, positioning Canadian Solar well for the next phase of our growth. Now let me comment on our guidance for Q1 and the full year 2013. We expect that Q1 shipments will be in a range of approximately 290 megawatt to 310 megawatt. We expect that Q1 demand will be mainly driven by U.S., Canada, Japan and Europe. We are also seeing strengths in India. We are well established in China but will move with caution in order to balance the opportunities and risk. European demand has improved lately but has the potential to become choppy due to the strong hints that the European Commission may consider implementation of radioactive duties on solar products manufactured in China. Gross margin in Q1 is expected to be in a range of 8% to 10%, showing further improvements from Q4 level. For the full year 2013, we expect module shipments of approximately 1.6 to 1.8 gigawatts, including modules used in our own project. Our forecasted shipment volume remains -- maintains our track record of continuous annual shipment growth. However, we have determined to set our path different from some of our competitors by avoiding profitless growth. Instead, our main focus for 2013 is to return to profit on annual basis. As you emphasize that, our guidance may be affected by both positive and negative market uncertainties, including the potential for higher growth in Japan and the potential for high growth in China but offset by the risk of delay of FIT payment flow or FIT reductions in this country. Implications of trade disputes between EU and China could also impact our results. We plan to mitigate these risks with further geographic diversifications while also taking extra precautions in production shipments and credit control. Overall, we expect global demand for solar project, as well as our shipments, will continue to grow in 2013, driven by U.S., Canada, Japan, China and other emerging markets. We also expect that nearly 50% of our revenue will come from our total solution business, driven by our utility-scale projects in Canada and U.S., as well as our residential system kits business in Japan. This, combined with the potential recovery of margin from our module business, should help us and help our business return to profit in 2013. This concludes my formal business update. I will now turn the call to Michael Potter, our CFO. Michael? Michael G. Potter: Thank you, Shawn. Net revenue for the fourth quarter of 2012 was $294.8 million, down 9.5% sequentially and down 37.8% compared to the year-ago period. Gross profit in Q4 was $14.9 million compared to $7.3 million in Q3 and $41.4 million in the comparable period of last year. The sequential increase in gross profit was driven by higher shipment volume and lower manufacturing costs. Our gross profit in the fourth quarter of 2012 is net of a $6.8 million depreciation charge, representing 2.3% of revenue related to the underutilization of our manufacturing assets. Gross margin in Q4 was 5% compared to 2.2% in Q3 and 8.7% in the fourth quarter of 2001 (sic) [2011]. Our gross margin came in above guidance primarily due to higher-than-expected ASPs. Operating expenses were $106.4 million in Q4 compared to $41.8 million in Q3 and $62.9 million in the comparable period of last year. The sequential and year-over-year increase in operating expenses was primarily driven by non-cash onetime charges totaling $61.2 million, including $31.2 million for doubtful accounts and $30 million for an arbitration decision against the company. We dispute the merits of the arbitration award and are continuing to evaluate our legal options. We are also hopeful that we will recover some or all of the doubtful accounts. However, it is uncertain today that we'll be able to do so, so we felt reserving the accounts now was prudent. Interest expense in the fourth quarter was $9.9 million compared to $15.2 million in the third quarter of 2012 and $11.7 million in the fourth quarter of 2011. The sequential and year-over-year decline in interest expense is due mostly to the change in the mix of our bank borrowings, with the higher portion related to the construction of projects. As a result, interest expense on our construction loans is capitalized under project assets and expense to constant revenue once project sales are recognized. In Q4, we recorded a gain in fair value of derivatives of $2.3 million compared to a loss of $5.3 million in Q3 and a gain of $2.4 million in Q4 of 2011. Net foreign exchange loss in Q4 was $10.8 million compared to a net foreign exchange gain of $7 million in Q3 and a net loss of $14.1 million in Q4 2011. Income tax benefit in the fourth quarter of 2012 was $3.3 million compared to income tax benefit of $1.8 million in the third quarter of 2012 and income tax expense of $16.3 million in the fourth quarter of 2011. Net loss attributable to Canadian Solar's shareholders for Q4 2012 was $105 million or $2.43 per diluted share compared to a net loss of $43.7 million or $1.01 per diluted share in Q3 of 2012 and a net loss of $59.9 million or $1.39 per diluted share in Q4 of last year. Excluding the onetime provision for the arbitration decision against the company and the bad debt provision allowance, adjusted net loss for Q4 was $43.7 million or $1.01 per diluted share. A table reconciling non-GAAP to nearest GAAP equivalent is available on our earnings press release issued earlier today and posted on our website. Moving on to the balance sheet. In Q4, cash and cash equivalents decreased to $142 million on December 31, 2012, from $319 million at the end of Q3. The decrease in cash was mainly due to operating results in Q4, the reduction of short-term debt inside China and the purchasing start of -- and start of construction of project pipelines in both Canada and the United States. The restricted cash balance was $422 million at the end of Q4, up from $371.9 million at the end of Q3. The increase was mainly due to the funding mechanism used to close and purchase the 4 projects in Ontario from SunEdison. Our accounts receivable balance, net of allowance for doubtful accounts, was down slightly to $254.9 million at the end of Q4 from $257.8 million at the end of Q3. Inventories decreased to $274.5 million at the end of Q4 compared to $317.8 million at the end of Q3. Inventories remain a focus area for us, and we hope to continue to make progress in this area. Short-term borrowings at the end of the fourth quarter of 2012 totaled $858.9 million, down from $890.6 million at the end of the third quarter of 2012. Long-term debt at the end of the fourth quarter of 2012 was $214.6 million compared to $224.2 million at the end of the third quarter of 2012. Let me take a minute to update you on our total solutions business. Our EPC utility-scale projects and solar systems kit business represented about 12.8% of total revenue in Q4 2012. In addition to our thriving residential system kit business in Japan, we expect the increased recognition of utility-scale projects in Canada and the U.S. to support the rapid growth of our total solutions business, which we will expect will account for at least 50% of revenue in 2013. Following the recently announced agreement to acquire 4 projects from SunEdison in Ontario, our Canadian project pipeline now includes 29 projects totaling approximately 400 megawatts DC. We estimate the resale value of our pipeline, once the solar power plants are completed, to be at least $1.5 million -- $1.5 billion. Starting with the 9 projects in Ontario sold to TransCanada. 2 of these projects, Brockville 1 and William Rutley, have been completed and connected to the grid in the fourth quarter of 2012 and, due to testing and permitting delays, announced to be -- now expect to be recognized as revenue in the second quarter of 2013. These plants are functioning and earning the feed-in tariff. As some of the very first for -- as they are some of the very first projects completed under the FIT structure in Ontario, we suffered from delays in the final permitting process, which we hope will not be repeated as more projects are completed and the process becomes better defined. 6 of the 9 projects sold to TransCanada are under construction, with Brockville 2 expected to be completed and recognized in revenue in the second quarter of 2013, positioning us to receive payment for a total of 3 projects in Q2. Burritts Rapids, 3 projects in New Liskeard, and Mississippi Mills are expected to be connected to the grid in the third quarter, with revenue recognition expected in either the third or fourth quarter of 2013. Alfred, the remaining project sold to TransCanada, is under REA, or Renewable Energy Approval, process and is expected to start construction in 2014. Among the 16 Ontario projects acquired from former SkyPower Limited, 2 have received REA approval and the remainder are expected to be approved in 2013. We expect to start building these power plants during 2013, with completion expected throughout 2014. Among the 4 projects recently acquired from SunEdison, 2 have received notice to proceed, which is the step after REA, 1 has received REA approval and 1 is awaiting REA approval. We expect to start construction on all of these projects in the next 12 months. In summary, of the 29 Canadian Solar-owned solar power projects in Ontario, 2 have completed construction, 7 are expected to be completed in 2013, 18 in 2014 and 2 in 2015. The company expects to recognize revenue using a percentage-of-completion method for all projects, except for the 9 sold to TransCanada. As a result, of the 18 projects expected to be completed in 2014, 15 will start construction and contribute to revenue in 2013. In addition to our owned projects in Ontario, we are building and plan to operate 3 solar power plants totaling 29 megawatts as EPC contractor on behalf of Penn Energy Renewables. These projects are expected to be completed in the second half of 2013. Turning to our U.S. project business. Our utility-scale pipeline now stands at approximately 255 megawatts DC. We expect up to 100 megawatts DC to be recognized in fiscal 2013. The project business certainly has its challenges to go with its rewards, but that is the reason why we built our initial pipeline with smaller projects in more mature markets. As we grow in experience and improve on our internal infrastructure, we expect to do larger projects and projects in more emerging markets. In summary, we are very pleased with the progress we have made in the transformation of our business from a pure module supplier to a provider of solar power solutions. As we move into 2013, Canadian Solar is well positioned with a strong brand, competitive cost structure and a solid pipeline of utility-scale projects to deliver a return to profitability in the quarters ahead. With that, I'd like to now open to your -- open the call to your questions. Operator?

[Operator Instructions] We have our first question. It's from the line of Kelly Dougherty from Macquarie. Kelly A. Dougherty - Macquarie Research: I just wanted to see if you could give us some idea in the mix between kind of straight module sales versus modules for the projects in the first quarter and in 2013 guidance? And then maybe given your expectations for when you plan to build the projects and recognize revenue, maybe a sense for how much revenue will come in the first half versus the second?

I will answer the first part of question, and then I will let Michael to answer the second part. This is Shawn. And for Q1, the Q1 guidance of 290 to 310 megawatt doesn't contain any utility projects. It will contain a little bit of the system kits business in Japan. Now as to the revenue recognition question, I defer -- I refer you to Michael. Kelly A. Dougherty - Macquarie Research: Shawn, can I have a real quick follow-up on that? How about in the full year, the mix between modules and projects from a megawatt perspective?

Okay. For the full year, we gave a guidance of 1,600 to 1,800 megawatt total, in which we expect the pure module sales to be around 1,300 to 1,400 megawatt, while the remaining, which is about 300 to 500 megawatt, will from the project business. However, on the revenue side, the revenue contribute from project business is going to be more than 50% of the total revenue. Michael G. Potter: In terms of the revenue recognition, and actually, we will recognize a small amount of modules from utility projects in Q1 for some percentage-of-completion contract projects we're doing, but it won't be that significant in Q1. The revenue for projects will be stacked more towards the back half of the year, a Q3 or Q4 time frame, mainly driven by the fact that the TransCanada projects that we use completed-contract methods, 5 of them are expected to be recognized either at the very end of Q3 or more likely than not in Q4. So that will tilt the revenue more into Q -- the second half of the year, that alone. Also, we're going to be ramping up construction of the projects that we bought off of the former SkyPower Limited throughout the year. So as we go through the year, more and more of them will be under construction and contributing revenue. So again, that's going to tilt the revenue more towards the second half of the year. Kelly A. Dougherty - Macquarie Research: That's helpful, guys. Then just a quick one on the cost. How should we think about the progression of the production cost this year? Obviously, you made a bigger per minute in 2012. So do you have a new target for how much lower your costs can go this year?

That's a good -- it's a good question. And this is Shawn. And as I -- as we mentioned, by the end of last year, we have achieved the $0.55 per watt all-in module manufacturing cost. And for this year, from pure processing point of view, we see somewhere around $0.02 to $0.05 further reduction, but that's on the pure processing side and because of the module is also affected by the polysilicon price, which we cannot accurately predict. Now assuming the polysilicon price and the resulting wafer price stay the same, then the module all-in manufacturing cost may trend towards $0.50 to $0.53 per watt range by the end of the year.

Next question is from the line of Satya Kumar from Crédit Suisse. Satya Kumar - Crédit Suisse AG, Research Division: I was wondering if we could talk a little bit about the gross margin potential of your total system business this year and next year? Michael G. Potter: So in 2013, the revenue is going to be dominated by our business in Canada, with some revenue contribution from our business in the U.S. Our Canadian projects are expected overall to average over 20% gross margin, with our projects we're selling to TransCanada to be over 25% gross margin. That's consistent with the guidance we've given in the past on that. Margins in the U.S. are lower than the 20% Canadian average. We do have some projects in other areas of the world, but they're not significant enough to discuss and not certain enough today to put onto our pipeline. They tend to range between 15% and 20% gross margin. Satya Kumar - Crédit Suisse AG, Research Division: How do you expect the OpEx for your business to track as you start getting -- recognizing these revenues from the utility business? Is the OpEx level more or less a function of your megawatts? Or will that start to scale as the utility revenues also start to increase? Michael G. Potter: There shouldn't be a significant rise in operating expenses because of the project pipeline. We've been building the infrastructure to support that for the last couple of years. There could be some adjustments in operating expenses going forward depending on what the different geographies of our revenues turn out to be, but that will be the natural turnover that you've seen in the industry over the last few years as solar markets have opened up and closed. Satya Kumar - Crédit Suisse AG, Research Division: Okay. And what were the details on the arbitration that you talked about that you took a charge for in Q4? Was that a purchasing contract only or something else?

This is Shawn. That's a arbitration decision made by the -- a arbitration board body in Shanghai regarding a contract dispute between Canadian Solar and LDK. And the result of that arbitration decision was in the news in December last year. And right now, the case is in the court system in Suzhou, City of Suzhou, and we continue to defend our position, we believe. We dispute the merit of this decision and will continue to defend ourself. Michael G. Potter: And it was for a wafer purchasing agreement. Satya Kumar - Crédit Suisse AG, Research Division: And one last thing. Michael, I think you said that you'd be -- or Shawn, I forget, but you're expecting to be profitable for the full year. Are you expecting to turn profitably in Q2 or, because of the revenue recognition being more weighted to second half, it's probably more a Q3 event?

Yes, this is Shawn. We -- at this moment, we are targeting the -- to be profit for the whole year. And as the quarter -- as the year goes on, there's better and better visibility for the profitability. And if everything goes our way, yes, we have a good chance for Q2. But however, at this moment, we are still looking at how -- when the first 2 projects which we built for TransCanada will go through all the permitting and testing and phase. So I will leave it, leave the answer to this question, to our Q1 Earning Call, probably in May.

[Operator Instructions] Next question is from the line of Nitin Kumar from Nomura. Nitin Kumar - Nomura Securities Co. Ltd., Research Division: A quick question on today's news regarding potentially a change in China's policy where they're talking about cutting feed-in tariffs to -- I mean, point at a [ph] CNY 0.75 per kilowatt hour, and I think the distributed number is also lower at 0.35 per kilowatt hour. I mean, how do you see it impacting demand? And do you think that the ASP -- lower ASPs in China will -- I mean, how do you see that impacting your gross margins?

Yes, this is Shawn. And as a matter fact, well, I haven't seen the news, today's news yet. However, we have heard those rumors in the past few months. There are lots of discussions of whether the feed-in tariff level in China will be reduced and -- but on the other hand, as everybody know, there's very little real cash payment for FIT from Chinese government yet. So, so far, this is only a policy. We haven't really see the cash payment flow yet. That's why, in our press release, we have cautioned the investor about the risk of business in China. And this is also why we have not given a guidance of high shipment in China. We are taking a cautious approach. And for this year, we are still going to focus our efforts to the markets which can give us profit rather than go for profitless growth. And as I -- as we said, this year's target is to return to profit. Nitin Kumar - Nomura Securities Co. Ltd., Research Division: Shawn, just switching gears here. I mean, a lot of your peers in China, like Yingli, Suntech, Trina, LDK, they have stretched balance sheets, focusing on China market. Now obviously, if you're not getting paid, the balance sheet gets even more stretched. Do you see these companies continuing to get bank support? Or do you think 1 or 2 may fall off the cliff sometime soon, or maybe it will in the next 12 months?

Well, that's a good question. And this is exactly why Canadian Solar has avoid this business approach. And we have budgeted very small module sales in China and we are developing pipeline -- project pipelines in China, but we are also very cautious in actually building any project in China. This is -- it's exactly because our desire to avoid those risks. And indeed, as you said, I -- we -- not -- many company have stretched their balance sheet by doing that. And -- but I can't -- I don't know, I can't really comment how the bank will react to their action. However, today, I did see the news from Bloomberg that Bank of China and -- or -- and a group of Chinese banks have started to sue Suntech for their bad loan. Michael G. Potter: And this is Michael. I could say that, towards the second half of last year, we increased shipments into the Chinese market to try and test the market and to see if we could establish reasonable and profitable growth there. And the bad debt expense that we recorded in Q4, about 2/3 of that is coming from Chinese customers. So it's quite possible that, over time, we'll recover a good portion of that doubtful receivables, but by our fairly conservative standards, we don't think it's probable or that we can say with great confidence today we can collect it. And that's one of the reasons why we're trying not to chase after growth where we can't assure collectibility in a reasonable time frame and we can't -- where we can't assure reasonable margins. Nitin Kumar - Nomura Securities Co. Ltd., Research Division: I mean, just a follow-up question on this. So if we assume -- I mean, basically, these companies are known as the big 4 or big 5, right? And if 4 of these companies don't have enough working capital, I mean, they can't stretch their balance sheets further, they would have to start cutting on their shipments elsewhere globally. Are you seeing demand gains -- faster demand gains in other markets? And if yes, do you have plans for capacity expansions for serving the remaining markets?

Well, at this moment, I don't have a plan for capacity expansion. Our strategy is to remain cautious. Especially, we remain cautious on pure module sales. For pure module sales, we'll only focus on low-risk countries and high-margin market. And meanwhile, we'll put our dollars into projects in total solution business but also focus on low-risk countries, such as Canada, U.S. and Japan. This approach seems to be conservative and it's clearly different from many of our competitors, but we believe that this approach is prudent, and we are -- and is -- we -- and is a responsible business approach.

Next question is from the line of Aaron Chew from Maxim Group. Aaron Chew - Maxim Group LLC, Research Division: Wondering if you guys could just offer an update on how utility-scale system pricing is trending across various markets. I mean, I guess you primarily have exposure to U.S. and Canada, so maybe you could focus on there. And if you have any additional insights into maybe how Europe and the rest of Asia is, that would be helpful as well.

Yes, this is Shawn. That will depend on the feed-in tariff offered by different countries. For example, the project we have in Canada was issued under so-called FIT 1.0, which guarantees a feed-in tariff of CAD 0.40 to CAD 0.43 per kilowatt hours for 20 years. So that's a very good feed-in tariff, so it can support a high project selling price. And for U.S., most of the projects without feed-in tariff is pure PPA and -- but with the investment tax credit, so that results in a lower selling price. Japan also have a high feed-in tariff, so the system price in -- the project price in Japan is also expect to be good. However, the -- there's also a discussion in Japan that the feed-in tariff level may be adjusted after the current fiscal year. So we are watching how it develop. So to answer your question, it all depends on the solar radiation condition and also the feed-in tariff level. Aaron Chew - Maxim Group LLC, Research Division: Okay, fair enough. And then just one follow-up. I wonder if you guys can comment on the prospective poly tariff coming out of China and how that may affect your wafer supply. I'm -- I know that you don't directly buy much poly. But if it seems that, that could potentially impact your wafer costs, or have you mostly figured out a way to get around that? Any color there would helpful.

Yes, well, as we said in the, yes, formal statement, that this is one of the uncertainties, one of the risks. And however, so far, we haven't seen much impact yet. Indeed, the polysilicon price increased a little bit. It increased from around $18 per kilogram to around $20 per kilogram, but this cost increase was more or less absorbed by the wafer producer. So we, as a cell module producer, haven't seen much impact yet. Now moving forward, we also have the wafer supplier and cell suppliers in -- outside China. So I believe we will have a reasonable approach to have wafer suppliers -- not really poly, but wafer suppliers using both China poly and oversea poly. So I think we have a strategy to respond to the potential tariff.

We have no further questions in the queue. Okay, we have another question now from Pranab Sarmah from Daiwa Capital. Pranab Kumar Sarmah - Daiwa Securities Capital Markets Co. Ltd., Research Division: My first question is on for 2013 outlook. Can you just remind me from which market you are -- what type of targets you are looking at, the different markets? Like from Canada and North America, how much of your total target, and how much will be from your Japan and other part of the world?

Are you talking about the total module shipment or only the project business? Pranab Kumar Sarmah - Daiwa Securities Capital Markets Co. Ltd., Research Division: Total module shipment targets, how you are dividing for 2013 outlook.

Oh, that's a very good question. But if you look at our Q4 geographic distribution, in Q4, about -- both Asia Pacific and Europe account for about 40%, respectively. North America is about 20%. Now moving into 2013, I see the contribution from Europe going down but the contribution from both Asia Pacific and North America going up, moving up. So as a result, well, North America and Asia Pacific will be the 2 leading areas while Europe will be -- probably be the #3. Pranab Kumar Sarmah - Daiwa Securities Capital Markets Co. Ltd., Research Division: And can you also give, by ASP differences, how it is looking at between Asia Pacific regions, especially China versus Japan and other part, and also North America's ASP prospects [ph]? Michael G. Potter: Yes, are you talking about module ASP? Pranab Kumar Sarmah - Daiwa Securities Capital Markets Co. Ltd., Research Division: Module ASPs, that's correct. Michael G. Potter: Yes, module ASPs are lower in China and India and fairly low in Europe and higher in other areas of the world. Pranab Kumar Sarmah - Daiwa Securities Capital Markets Co. Ltd., Research Division: The lowest is now in Europe, then followed by China and India. And after that [indiscernible] U.S....

No, the lowest module price have been in India, China and Thailand. And Europe is a little bit high -- is higher, and U.S. is a little bit more higher. And in Canada and Japan, we have higher price. And then in some emerging market, the price is even -- now you see, it's much better, but those emerging market doesn't demand much volume. And however, the situation may also change. For example, Japanese yen have went down almost 20% in the past few -- a couple of months, and that results in the profit margin in Japan shrink significantly. But still, Japan is one of the high-margin market for us. It's also because, in Japan, we adopted a brand strategy. Rather than selling OEM, we sell Canadian Solar brand in Japan, so we had a chance to sustain a good price. And also in Japan, we -- for the residential market in Japan, we deliver the complete kits rather than deliver only module. So by providing the roof -- residential roof solutions, we also sustain a reasonable price in Japan.

[Operator Instructions] Next question is from the line of Timothy Lam from Citigroup. Timothy Lam - Citigroup Inc, Research Division: My question is related to company's plans for the shipments into 2013. Given that the market is quite strong in Japan and in China, it -- what kind of percentage of sales the company anticipates to sell to these markets as the overall in view of that? I know that -- secondly, a follow-up question to this is, I know that company has not been engaging too much into the Chinese development projects, but does this see any opportunities to target some of the projects that are happening in Japan?

Well, I can give you a snapshot of what we see at this moment. At this moment, out of the forecasted 300-megawatt shipment in Q1, I see around 1/3 goes into North America. And then I will say about 1/3 goes to Asia Pacific and Europe is a little bit lower, and then the rest goes into other part of the world. So you may use this number as an indication for the full year geographic distribution. Michael G. Potter: Yes, and commenting on the China projects and Japan projects. We're very active in the Chinese project market and we're building what we believe will be a very solid pipeline for the future, but we're not engaging in building out large projects in China until the market is more certain, until the end buyers and the process to find end buyers is more certain as well. So like our module sales, we're pursuing profitable growth, not profitless growth. In terms of Japan, we've been in Japan for quite a long time and we're actively building a pipeline there, and we believe we'll have good news to disclose in the future from our results there.

Now I will add one statement. As you all know that I'm a China-born Canadian citizen, so I know China very well. I believe in the Chinese -- China market. I believe that this market will eventually -- well, this year, China market may become the biggest, but it's not going to be the -- a good market because of the -- there's almost no profit, and also the credit risk. But into the future, China may become a big but also solid market. So that's why we are preparing for this market, but I'm -- I prepare to -- only prepare the market for the growth in 2014 and 2015. Michael G. Potter: And I'm not a -- as people who've met me know, I'm not a Chinese-born Canadian, I'm a Canadian-born Canadian, I guess. And I also strongly believe the market in China is going to be very strong in the next couple of years. And if you go back and look, even the Canadian market, there was a lot of delays in the rollout of the market and a lot of uncertainty. And it took a couple of years before the government policies and the desire to do solar started turning into solar power plants being built. So China is no different. There's only a lot more people and a lot more opportunity here, but it's going to take a while to work from policy into action, into a system that works well.

Yes, and I want to further echo that, what Michael said. For example, Canada, Ontario issued its Green Energy Act in 2009. So we submit the project. In early 2010, they issued a first round of FIT approval, but only by the end of 2012 the REA got start to get released, so the project only get to be built in late 2012 and then 2013. That shows how much time a new market will take to get developed. So China is no exception. It's somewhat a belief that a China project market can turn into a mature market overnight. Personally, I think, not agree. So our strategy, again, is to focus on growth, but focus on growth with profit and also focus on low-risk countries.

We have no further questions in the queue. I would now like to turn the call back to you for any closing remarks.

Thank you, operator. And thank you, everyone, for joining the call today, and thank you for your continued support. If you have any further follow-up questions after today's call, please contact us. And have a great day. Thank you.

Ladies and gentlemen, that concludes your call for today. Thank you for joining. You may now disconnect.