Canadian Solar Inc. (CSIQ) Q1 2012 Earnings Report, Transcript and Summary

Good day, ladies and gentlemen, and welcome to the Canadian Solar First Quarter 2012 Earnings Conference Call. My name is Deanna, and I'll be the 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 your host for today, Mr. David Pasquale, Global IR Partners. Please proceed.

Thank you, operator. Welcome, everyone, to Canadian Solar's First Quarter 2012 Earnings Conference Call. With us today are Dr. Shawn Qu, Chairman and Chief Executive Officer; Mr. Michael G. Potter, Senior Vice President and Chief Financial Officer; and Mr. Ed Job, Director of Investor Relations. Before we begin, please be advised that management's prepared remarks and responses to your questions during this call may contain forward-looking statements that involve a number of risks and uncertainties that could cause actual results to differ materially from results anticipated by these forward-looking statements. These forward-looking statements are made under the Safe Harbor provisions of the U.S. Private Securities Litigation Reform Act of 1995. You should not place undue reliance on forward-looking statements, and we encourage you to review a more detailed discussion of the risks and uncertainties contained in the earnings press release issued earlier today and in the company's annual report on Form 20-F filed with the Securities and Exchange Commission. All information provided on today's call is as of today's date, unless otherwise stated, and Canadian Solar undertakes no duty to update such information except as required under applicable law. At this time, I would like to now turn the call over to Dr. Shawn Qu. Please go ahead, sir.

Thanks, David, and thank you, all, for joining us on today's call. We are pleased with our results for the quarter. Shipment of 343 megawatts and average selling price of $0.89 per watt were in line with our guidance. Demand was even higher, but we could not fulfill every order due to the Chinese New Year holiday. That said, we also remained selective in pursuing opportunities most aligned with our profit objectives. That discipline, combined with our ongoing cost initiatives, helped us deliver gross margin of 7.7%, which is at the top end of our prior guidance of 5% to 8%. The key theme for Canadian Solar in the past few quarters has been solid execution at all levels of our organization. Our R&D continues to push forward with exciting new advances to improve the efficiency, the look and functionality of our product and to reduce the total cost of ownership. As an example, in the first quarter, commercial efficiency of our monocrystalline ELPS cells reached 19.5% in mass production, while in the lab, ELPS version 2.0 reached 20.5% of conversion efficiency. Our sales team under Yan Zhuang, our newly named Chief Commercial Officer, has done a tremendous job as we have consistently met our shipment guidance through this challenging macro environment while, at the same time, increasing market share. We have become the #2 largest PV module producer in the world in Q4 2011. During Q1 of this year, we believe that we have maintained and further strengthened our status as one of the world's top-ranking manufacturers while further diversifying our geographic footprint. We are also very pleased with our manufacturing and operations team under Charlotte Xi, our Senior VP for Operation. Through their consistent and tireless efforts, we have reduced our blended manufacturing costs to $0.33 per watt in the first -- sorry, to $0.73 per watt in the first quarter compared to $0.84 in the first quarter of 2011. To be clear, our blended manufacturing cost includes purchased silicon, cells and wafers but excludes warranty provision, as well as impact of high-cost inventories carried from the previous quarters. And finally, our finance and accounting team under the leadership of Michael Potter has become another differentiator for Canadian Solar. Michael's team has helped to strengthen our internal systems and controls at an important time, given current competitive environment and our expansion drive in the PV total solution business area. Not only have they helped arrange financing for our own project, they also have delivered innovative financing solutions to our customers. We are now a partner of choice for customers in top solar markets worldwide. Importantly, our strengths in sales and marketing, manufacturing and financing are helping us maintaining strong cash and financial conditions while positioning ourselves to emerge from this cycle as a stronger leader of the PV industry. We're excited about our business and growth opportunities. We expect Canadian Solar to continue to benefit from the global flight to quality. We also expect that as the year develops, investor will start to afford us a more appropriate valuation that recognize our consistent execution, transparency, growing market share, differentiated business model and a higher visibility of our rapidly growing total solution business. Now let me give you an update on our total solution, our PV project business. We expect to collect payment in Q2 on the last of our 3 solar power plant EPC jobs, which we started in 2011 in Ontario, Canada, each with 10-megawatt AC. We have started construction of 3 other 10-megawatt AC projects in Ontario, which we own ourselves and we will build and transfer. One of them is expected to recognize revenue in Q3 as we connect it to the grid and sell it to off-takers. And the other 2 are expected in Q4. As recently announced, we have secured construction financing from Bank of China to fund these projects. In addition, we expect to receive full permitting of 5 to 6 more projects and start construction at different times throughout 2012, with completion in 2013. These projects will also benefit from construction financing provided by the Bank of China facility, as well as other funding sources our financing team is working on. We expect these 5 to 6 projects and their permitting process and also the 16 project, which we recently acquired from SkyPower, to be built and sold through 2013 and 2014. Taken together, this means, on average, about 3 projects turn into revenue every quarter in 2013 and 2014, which provides us with a visible and a captive total solution business. We started our major project business in our home province of Ontario, Canada, but we are rapidly building a pipeline elsewhere as well. We expect to discuss our project pipelines in U.S., China, Southeast Asia in great details next quarter, but I want to identify this as another focus area for us. The reason today's passed feed-in-tariff in Japan is also expected to open up the country for utility scale projects. We expect to benefit from that as we leverage our long successful track record as one of the Japan's market leaders. You cannot execute on project without arranging financing. Canadian Solar invested in project financing team several years ago specifically because of this. That investment in people and banking relationships is now being realized. Geographically, the majority of our current projects are in Canada, with the remainder in U.S. and many potential projects in various development stages in China, Southeast Asia and Europe. The expanded screen of Canadian Solar in the project area was again seen last month when we announced a landmark deal with SkyPower. Under this agreement, Canadian Solar will acquire the majority interest in 16 solar projects, representing approximately 190- to 200-megawatt DC from SkyPower. The size of each individual project is in line with our strategy of concentrating our smaller utility scale projects, which is less risky to finance and to execute on. These projects acquired from SkyPower are expected to generate over USD $800 million in revenue for Canadian Solar as we successfully complete the construction and sale of each project. We are also excited about the related joint venture we announced. This is a 50-50 international JV with SkyPower focusing on developing solar power plants in select emerging market. We expect to start generating revenue from the JV over the next 2 to 3 years. When you take our latest transaction with SkyPower on top of our Q4 deal to sell 9 utility scale projects to TransCanada, you can see our momentum and our credibility in our goal to expand our total solution business, which consists of project development and sales, EPC and solar system kits. We would like to remind you that we have publicly announced a 3-year target back in 2010, which is to bring the total solution business revenue to 10% of our total revenue in 2011, 25% in 2012 and over 40% in 2013. We are on track to reach these strategically important goals. Now let me comment on our guidance for Q2 and the full-year 2012. We expect Q2 shipments will be in a range of approximately 430 megawatt to 450 megawatt. We expect sales to come in from all of our key markets, including Germany, Italy, France, U.K., Japan, U.S., Canada, China, India, among others. Gross margins are expected to be in a range of 8% to 10%. I also expect that we will continue to execute our cost-style roadmap and bring our blended module costs to reach $0.50 to $0.60 per watt before the end of year. Overall for 2012, we reiterate our prior guidance for shipments of approximately 1.8 to 2 gigawatts. That concludes my formal business update, and now I will turn the call to Michael Potter, our CFO. Michael, please go ahead.

Thank you, Shawn. I'll review a few points before turning to Q&A. Please refer to our press release for additional detail. Net revenue for the first quarter of 2012 was $325.8 million, down 31% sequentially and down 26.5% compared to the year-ago period. Gross profit in Q1 was $25.1 million compared to $41.4 million in Q4 and $65.3 million in the comparable period of last year. Gross margin in Q1 was 7.7% compared to 8.7% in Q4 and to 14.7% in the comparable period of last year. The sequential and year-over-year decline in gross profit margin was primarily due to the sustained decline in average selling prices over the past several quarters, which is only partially offset by lower manufacturing costs. I'd like to note that the Q4 number of 2011 included onetime items. And without those onetime items, we actually increased gross margin sequentially into Q1. Operating expenses were $38.5 million in Q1 compared to $62.9 million in Q4 and $31.1 million in Q1 of 2001. Recall in Q4, there was a provision of $17.4 million against a prepayment made for a long-term wafer agreement, and we also had higher costs associated with supporting our higher shipment levels. We continue to evaluate all costs organization-wide. We want to ensure that we're deploying our resources in the right markets and against the right opportunities. Given the current operating environment, it is clear that strict financial discipline is the differentiator which will separate the winners. CSI plans to be a winner as we build on our already market-leading position through further market share gains. In line with our efforts to evaluate our costs, we've worked to institute across-the-board reduction to a termination of service providers of CSI, and we've reduced non-essential headcount where possible. Because we've always run leaner on operating expenses, these reductions are not large, but we continue to reexamine where we spend our operating expense every day just as our factories squeeze costs. Our actions are efficiency measures and have not impacted our ability to provide exit sales and aftermarket support to our customers. As we also mentioned in our press release, in Q1, we recorded a slight loss from the change in fair value of derivatives of $0.2 million compared to a gain of $2.4 million in Q4 and a loss of $17.2 million in Q1 of 2011. Foreign exchange gain in Q1 was $0.3 million compared to a foreign exchange loss of $14.1 million in Q4 and a loss of $0.7 million in Q1 2011. In Q1, income tax benefit was $2.5 million compared to an expense of $16.3 million in Q4 and an expense of $1.7 million in Q1 of 2011. Net loss attributable to Canadian Solar shareholders for Q1 2012 was $21.3 million or $0.49 per diluted share compared to a net loss of $59.9 million or $1.39 per diluted share in Q4 of 2011 and net income of $5.9 million or $0.13 per diluted share in Q1 of last year. Moving on to the balance sheet. In Q1, cash and cash equivalents increased to $393 million on March 31, 2012 from $344 million at the end of Q4. We generated approximately $12.1 million in cash flow from operations during the quarter. The restricted cash balance was $231.8 million at the end of Q1, up from $178.3 million at the end of Q4. The increase primarily resulted from the cash collateral for bank borrowings. Our accounts receivable balance net of allowance for doubtful accounts decreased to $251 million at the end of Q1, down from $292 million at the end of Q4. The 14% decline is in line with the sequential decline in revenues. Inventories increased to $389.9 million at the end of Q1 compared to $296.6 million at the end of Q4. Inventories remain a focus area for the company, and we continue to make good progress. The sequential increase was related to an increase in work-in-process and finished goods and also related to the timing of the FIT change in Germany, which delayed some orders in March, which have since shipped in April. Let me take a minute to update you on our project business. Our EPC and solar systems kits business represented about 7% of total revenue in Q1 2012. As Shawn noted earlier, we expect this to grow to 25% of revenue in 2012 and 40% of revenue in 2013. Let me discuss our shorter-term Canadian pipeline in more detail. We're currently constructing 3 Canadian Solar owned projects. One is a RESOP and 2 are FIT contract projects that have TransCanada as the end buyer. Additionally, we have a pending REA for FIT projects, with a target of Q3 2012 for the start of construction and engineering currently in progress. These are 2 projects that we expect to sort of hit that milestone. We also have 4 pending REAs FIT projects that are targeted to start construction in Q4 of 2012. We have 1 final project to submit for REA that we expect to start construction some time in 2013, and that makes up to 9 FIT projects we announced in our TransCanada announcement. For the owned projects I mentioned, we'll be using the completed contract message to recognize revenue. Typical revenue recognition will be 2 quarters after we start a construction, and the first of the projects I discussed is expected to be completed and paid for in Q3, with the remaining 2 of the 3 that will start construction in Q4. We have other projects in Canada in our pipeline where we're applying to be the EPC contractor, similar to the 3 projects we completed in 2011. As Shawn mentioned, we have an even broader pipeline outside of Canada that we'll discuss in more detail later this year. In summary, Canadian Solar continues to successfully execute on our business strategy. We are fully committed to expanding our leadership position, enhancing our profitability and delivering meaningful value to our shareholders. We believe we're in a great position to build on our momentum and our clearly differentiated business model. With that, I'd like to turn over the call for your questions. Operator?

[Operator Instructions] The first question will come from the line of Kelly Dougherty, Macquarie.

I just want to follow up on the systems business a little bit more. Within that 25% contribution to revenue in 2012, can you help us think about how much of that is actually the full project work that you're just talking about versus just EPC work versus selling the kits in Japan and maybe how we can try to work that into the model?

The majority of the revenue this year is going to be from projects that we own. Although we will enjoy close to about the same revenue this year as we did last year from EPC projects. And almost all of the revenue expected from Japan is in the form of full kit. So geographically, revenue from Japan are residential kits, and those will be counted in that number as well. Yes, we have more projects that we own that we'll be completing just in Canada, and we'll talk about that in more detail later on this year.

Okay, that's helpful. But can you give us any kind of sense of what ASPs look like for the 3 different buckets within that systems business? And then just a quick follow-up, within the 1.8 to 2 gigawatts, is that production or shipments? And I wonder if any of that'll be deferred given the project work.

Well, I can give you a rough approximation. We said we'd sell 9 projects for approximately USD $450 million. So that would be about $50 million per owned project as an approximation. We did 3 projects, EPC business, for approximately $100 million last year. And the systems and kits business, you can get from looking at the Japan revenue from prior years. In terms of the 1.8 and -- to 2 gig of that shipment, it's not produced. Recognition -- owned projects are using the completed contract method. If we're doing EPC work, we're probably going to be using percentage of completion and recognizing it as we do the work.

So how much do you expect to recognize from a megawatt perspective this year?

For these projects, I mean, the average project in Canada is about 10 megawatts to AC, somewhere around 12 megawatts DC. So in terms of megawatts, it's not a very large number. In terms of revenue, it's much larger contribution.

And the next question comes from the line of Satya Kumar, Crédit Suisse.

I was wondering if you could comment a little bit more about the pushouts that you mentioned that happened late in the quarter in Germany. Can you give a sense of the magnitude of the orders that were delayed? And specifically, what was different between what your customers were expecting and what happened in Germany? And when do you now expect these FIT tails to take effect, and how that might affect your shipment to Germany?

Yes, I will -- this is Shawn. I will pick up this question. Early this year, German, the politician put out a draft, which called for a change of feed-in tariff policy on March 31. But later on, close to end of March after the many negotiations, finally, the German Parliament made a decision to have that policy switched on June 30. So the deadline for the policy change was pushed from March 31 to June 30. Therefore, there are some product, which, in the past, we thought we will have to ship out in March. We actually shipped them out in April. But all those products are already shipped, and you will see it in the Q2 revenue.

I mean, the simple answer was is that German market paused for a little bit. Well, they adjusted what the actual FIT number would be. And then once it was obvious that what the new dates were, we got a decent number of orders that wanted delivery in April and early May as well. So we staged some inventory in Q1 when our production lines were a little bit emptier so that when we entered Q2, we'd make sure we have sufficient supply when the German market needed it.

And the next question comes from the line of Pranab Sarmah, Daiwa Capital Markets.

My first question is on those solar farm project. Could you give us a little bit of color on what type of profitability you are looking for those projects this year and probably the following year?

When we talked about the gross margin on the projects last year when our margins were in the mid-teens, we said there was about 10 points higher than that on average. So the owned projects in Canada we expect to be quite nicely profitable. EPC has slightly lower profitability, and the profitability in other areas of the world are not quite as high. The reason in particular the 9 owned projects are profitable for us is that we develop them pretty much from the ground floor up. We are a Canadian company founded in Ontario. So when the Ontario government decides to enter into the feed-in-tariff program, we were there and able to apply for and develop projects from the beginning.

[Operator Instructions] The next question will come from the line of Hari Chandra, Auriga USA.

This was more from a longer-term standpoint. Looking at the trajectory of your debt-to-capital ratio, it looks like most of the growth seems to be debt-induced. So my question is at what point would you consider debt a drag and being value destructive? And as much as you're executing on the side, there seems to be financially -- these things do not seem to be profitable for the company.

So last year, we actually reduced our debt significantly at the end of the year as we had cash flow that's geared towards that. This year, the debt that we're incurring is going to be shorter-term debt to construct the projects and then later sell them and pay down that debt or a debt to finance the SkyPower acquisition, which again is going to be turned into project revenue. I think my debt metrics compared to just about anybody else in the industry are quite low. We're comfortable with the levels that we have now and what we need to do to develop and execute on our project pipeline. Long-term, the goal is to change more the short-term debt into longer-term debt and to pay down debt where we have the opportunity to.

Yes, this is Shawn. I would like to add another comment. There's no doubt that solar industry is in a consolidation phase right now. However, Canadian Solar have focused this consolidation phase at least 1, 2 years ago. So ever since late 2010, Canadian Solar has been executing on one hand a conservative CapEx strategy and on the other hand, we have been focusing on technology development and also the change of business model from pure module cells into project business. And as of today, on the pure module cell side, Canadian Solar have been able to increase our market share even in the market -- even in the industry downturn, and we have maintained gross margins in every quarter so far. Now meanwhile, we have put -- we have invested into project business in the past 2 years, and now we start to see this effort pay back.

And the next question comes from the line of Colin Rusch, ThinkEquity.

Can you talk about the introduction of AC modules, and when you expect to have those products in the market in volume? And also what your supply chain looks like on the [indiscernible] stuff?

Yes, this is Shawn. We have a plan to introduce AC modules. We introduced, we communicated with market on this plan in late last year. At this -- right now, we have demo systems running in several continent in both Asia Pacific, Germany and also in North America. And we expect to start to introduce AC module product either late this year or early next year. We decided to do a little bit longer testing period. Now, on the other hand, we -- I expect the -- just the high efficiency DC module in our total solution business will be the main contributor to our revenue and profitability, both for this year and for next year.

Great. And within the Americas sales that you've had, can you break out anything that's not in Canada and North America at this point, and what your expectation is for the balance of the year?

In Q1, most of the revenue that says America actually was in the U.S. We didn't have a lot of revenue in Canada in Q1.

Yes, in Q1, Canada is still a winner, and -- but we are working on the -- we are constructing a few projects. But due to the fact that we used the completion method for revenue recognition, you'll only see the revenue of those efforts when we connect the project to the grid and sell them to the off-taker.

And the next question comes from the line of Ahmar Zaman, Piper Jaffray.

This is Karen on behalf of Ahmar. Congratulations on the impressive cost reduction from $0.84 to $0.73 in 1 quarter. Can you elaborate on what specifically contributed to the cost reduction, maybe a breakdown of your wafer costs in the quarter, as well as your cell and module processing costs?

Yes, this is Shawn. The contribution come from all areas. On the wafer side, the -- we had about $0.02, $0.03 cost reduction, and I think when we started the quarter, the wafer costs worked out to be somewhere around $0.30 per watt also. And when we exit the quarter, it become $0.26 per watt also. It's due to the wafer price reduction but also due to our -- the increase of the wattage, the increase of the cell efficiency we achieved on the same wafer. Now on the cell side, we also achieved a significant cost reduction, about $0.02 to $0.03, to reduce the cell processing costs, again, to increase efficiency, therefore, we produced more wattage out of the same material, and also to increase the yield on the cells. Now for Q1, the most significant cost reduction come from the module area, and we -- and it's with our efforts to increase the output and also to increase the automation level, weighing process to automate our solar module workshop in China. And as you know, even in China, the wage increase has been the topic in the past 2, 3 years. And now we have successfully bring the cost of automated workshops below because of the manual workshop. So altogether, that gave us over $0.10 contribution. And also automation that we are seeing the same trend. It's not a bottom. We are seeing the same trend in Q2, and I'm very confident. I want to say I'm very confident that we will continue this trend, and we will definitely bring the blended module costs in Q2 below $0.70. That's why we are saying that we are very confident that we will execute our plan to achieve $0.55 to $0.60 OEM blended module costs before the end of the year.

And the next question comes from the line of Aaron Chew, Maxim Group.

Just wondering if you guys could comment or provide some color on system pricing. So I understand that you can't speak to any specific contracts. Perhaps you could just speak to the extent to which system pricing is keeping pace with the module price declines over the last year. Where are system prices now? Have they declined this year? Are we firmly below $3 a watt or USD $3 a watt? A minimum, maybe you can talk about Ontario. I don't know if you can just said shed some additional insight you may have into the German, U.S. or Asian markets as well.

Yes, I want to make a quick comment, then I leave it to Michael, my number guy. And on pure EPC, the EPC, pure EPC quote also come down as the module costs come down. In some places, we have seen module plus EPC below $3. For example, you mentioned in U.S. and Canada, yes, in those places, you'll start to see below $3. But that's only on the hardware like module EPC and the installation. It doesn't include the soft costs like permitting, engineering design, all that stuff. That stuff will add up and will depend on different markets. Now in emerging market, the situations are different. There are several quote in some emerging market, which is between $3 to $4. And then sometimes we also quote our off-grid project. For those projects, the price is really case-by-case. Now this is on EPC system costs. Now as you noted, that our total solution business also include rebuild and sell, flip our own project. In this case, when we build, sell and flip our own project, then the investor only look at the IRR, only look out how much IRR they can -- whether they achieve the IRR target. It really doesn't matter what is our EPC cost. Now, Michael, do you have any comment?

You said that you'd passed it over to the numbers guy, but you went through the numbers pretty well. The pricing really depends if you have sold like a completed system and you're selling it, the end selling price really ties into the feed-in-tariff for the PPA and the period of time and the financing costs and any the tax advantages in the individual country. So basically, with the drop in the module costs over the last year or 2, anybody that's had any sort of a loss in PPA or FIT, the developers made out fairly well because the actual cost of construction went down, but the selling price didn't really change. So it really depends on which market. Canada, the older FIT projects happen to have a higher selling price because the FIT is higher. The new FIT that was announced, we expect to have a lower end selling price for the systems. Of course, the costs have come down, so they're still profitable. In countries like the U.S., the funding environment, the cost of funds is less expensive. So the IRR expectations for the end buyers are also less. So the prices vary according to that. Emerging markets where there may be country risk or other sort of political risk, investors tend to want a higher return. So that actually pushes down the price of the projects a little bit. It just depends on which market. I can't really give you specific prices. And the younger it is in solar for our country, the more varied the prices are. In markets like Germany that have been doing solar projects for quite a long time, the prices are much more standardized.

And the next question comes from the line of Mark Bachman, Avian Securities.

Michael, can you talk about your $0.73 manufacturing cost in the quarter? It looks like you fell short. You had an idea that you would get to $0.70 there. So maybe if you can break down wafer selling module for us and tell us why you fell short there by $0.03.

Yes, that's the blended cost for the entire quarter. We said we'd reached $0.70 by the end of the quarter. So we didn't fall short. So in terms of that, we're fine. But for the cost, the blended cost quarter-over-quarter, there was a big reduction. Like Shawn said, we're targeting $0.55 to $0.60 at the end of the year, not for Q4 blended costs. And at the trajectory we're on right now, we're confident of those targets.

Yes, I also want to add a comment that late January, early February is the Chinese New Year, and we have over 10 days of holiday or factory shutdown. Of course, the cost in those 2 months is higher since we still have carry -- pay the salary to our employees. And in Q3, once we turn the factory to a full speed, then the costs really start to come down.

[Operator Instructions] And the next question comes from the line of Kelly Dougherty, Macquarie.

Obviously, with the overcapacity in the market right now, it's kind of taboo to talk about adding more. But if you're thinking of selling 1.8 to 2 gigawatts this year and current module capacity, correct me if I'm wrong, I think it's about 2.1, how do you look into 2013 from a capacity standpoint? And if you add anything, would it just be module as you have the partnerships on the cell and the wafer side?

Yes, this is Shawn, Kelly. With our current capacity, we can deliver our 1.8 to 2 gigawatts without problem. As we said in the press release in Q2, we expect to deliver about 430 to 450 megawatts. With this speed, we will be able to reach 1.8 to 2 gigawatts. And right now, we are looking into some expansion plans. On the module area, we are looking at another automation -- automation of another workshops to improve the efficiency and further cut down the number of workers. Now we haven't made a final decision yet, but for that project, we already have the bank financing lined up. So if we decided to go ahead, then the CapEx financing is there. And on the cell area, we are considering a plan to move with our ELPS 2.0. That's our new technology, which give us 20.5% cell efficiency, which is about 8% to 10% higher than the normal cell in the market. So we believe that will differentiate us from a technical side. So we want to differentiate ourselves not only by the share, market -- share size and cost but also by technology and by total solution business model. So on the cell area, we do consider one project, but we haven't reached a final decision yet. I believe by the next earnings call, I will be able to update you. And of course, Michael and I would like to watch how the company fare in Q2, Q3 and then to be sure that we have sufficient demand for our high-efficiency solar module.

And the next question is a follow-up question from the line of Pranab Sarmah, Daiwa Capital Markets.

My question is on Q1. What percentage of your shipment were on high-efficiency module out of that 460 megawatt?

In Q1, out of our 343 megawatts?

Yes, I will answer this question. I believe -- yes, this is Shawn. I believe it's about 5, 6 -- 4, 5 megawatts also because with a high-efficiency ELPS cell, right now, we are running full speed on one line. We are accumulating experience right now. So that line produced about 3 -- 2 to 3 megawatts a month.

Okay, got you. And, Shawn, could you also give some color on the industry consolidation? As you know, like China, 12-year plan phase, like they want only 10 strong solar company in China. How do you view this situation from the government, and how Canadian Solar is preparing for such situation?

Yes, this is Shawn. We really have no comment. First of all, we are a Canadian company. We manufacture in China, but we are a Canadian company. So we are not really governed or bound by whatever ministry decree or plan. The second, you see this kind of plan pretty often. It's more or less, I will call it, an expectation and -- but industry is industry. We are private-owned. Most of the solar industry are private-owned here. It's not that the government issue a decree, something will happen. So I think we will just do a good job ourselves and eventually, competition will, say, -- will show who is the winner in this consolidation phase.

[Operator Instructions] And we have a question on the line from Kelly Dougherty, Macquarie.

Just one more for me. Michael, can you tell us when you think you'll be through the higher cost inventory, such that your cost of goods will better reflect the $0.55 to $0.60 or the $0.70 target that you talked about for 1Q and for the full year?

Yes. So I would guess when we stop driving costs down or have 0 inventory from the prior quarter, the idea is to control the inventory down, which this quarter will be easier because our shipments are higher. We still need some inventory just looking at the goals and looking what we have to do in the second half of the year. You have to take advantage of lower shipment months to build up a little bit of a buffer. But as long as the prices continue to decline, the inventory from the prior quarter is going to be higher-priced inventory than the quarter before. We do not have very old inventory that from 2 or 3 or 4 quarters ago, that are very, very, very high prices. So it's higher priced. It came from the prior quarter, but it's not like super high-priced.

Yes, Kelly, this is Shawn. I will add a few words. Usually, you see half-month or 1-month inventory carried from previous quarter to the current quarter. So if the -- for example, in Q1, the blended cost for Q1 is around $0.73, but we carry some inventory from Q4, so I think our cost of goods sold is around $0.78 also. So it's a few cents higher. That reflect the inventory we carry from Q4. Q2 is probably going to be the same because the ASP went down from Q1 to Q2. And so if ASP continue to go down quarter-by-quarter, then I guess we will typically see 15 to 20 days of old inventory being carried through the current quarter, but as -- so mostly, it's 1 quarter. In 1 quarter, you get evened out. And if the price and costs start to stabilize, then, of course, we don't see this phenomenon anymore.

And ladies and gentlemen, this concludes the question-and-answer portion for today. I would now like to turn the call back to Canadian Solar's Chairman and CEO, Shawn Qu.

Thanks, operator, and thank you, everyone, for joining the call today. And thank you for your continued support. Michael will be attending the Deutsche Bank Clean Tech Conference in New York on the 15th of May. So if you would like to schedule a one-on-one meeting with him, please contact the DB corporate access team. And if you have any further follow-up questions after today's call, please contact us. Have a great day.

And thank you again, ladies and gentlemen, for your participation. This concludes today's conference. You may now disconnect, and have a great day.