第一太陽能 (FSLR) 2017 Q2 法說會逐字稿

Good afternoon, everyone, and welcome to First Solar's Second Quarter 2017 Earnings Call.

This call is being webcast live on the Investors section of First Solar's website at firstsolar.com.

(Operator Instructions) As a reminder, today's call is being recorded.

I would now like to turn the call over to Steve Haymore from First Solar Investor Relations.

Mr. Haymore, you may begin.

Thank you.

Good afternoon, everyone, and thank you for joining us.

Today, the company issued a press release announcing its second quarter 2017 financial results.

A copy of the press release and associated presentation are available on the Investors section of First Solar's website at firstsolar.com.

With me today are Mark Widmar, Chief Executive Officer; and Alex Bradley, Chief Financial Officer.

Mark will provide a business and technology update, then Alex will discuss our second quarter financial results and provide updated guidance for 2017.

We will then open up the call for questions.

Most of the financial numbers reported and discussed on today's call are based on U.S. generally accepted accounting principles.

In the few cases where we report non-GAAP measures, such as free cash flow, adjusted operating expenses, adjusted operating income or non-GAAP EPS, we have reconciled the non-GAAP measures to the corresponding GAAP measures at the back of our presentation.

Please note, this call will include forward-looking statements that involve risks and uncertainties that could cause actual results to differ materially from management's current expectations.

We encourage you to review the safe harbor statements contained in today's press release and presentation for a more complete description.

It's now my pleasure to introduce Mark Widmar, Chief Executive Officer.

Mark?

Thanks, Steve.

Good afternoon, and thank you for joining us today.

Our operational financial results for the second quarter were resilient and market demand for our technology continues to be robust.

We've had a number of highlights since our last earnings call, including the arrival of our first Series 6 equipment at our factory in Ohio, record quarterly shipments of nearly 900 megawatts DC and strong bookings of 1.5 gigawatts DC.

In addition, our financial results for the quarter were solid, with non-GAAP earnings per share of $0.64 and an ending net cash balance of $1.9 billion.

The sale of Switch Station and higher module sales to third parties were important contributors to the quarterly results.

As a result of our first half performance, improved visibility and systems project sales and an increase in expected shipments, we have raised our full year 2017 net sales, EPS, operating cash flow and net cash guidance.

Entering 2017, we knew it would be a challenging year for us with the uncertainty of the global supply/demand balance and our product transition to Series 6. Despite these challenges, we have made great progress in the first half of this year.

Firstly, following the receipt of a waiver under the ROFO agreement with 8point3 for interest in the Switch Station project, we were able to leverage the continued vigorous market demand for our high-quality systems projects and sell the interest in Switch Station at a significantly higher valuation versus selling to 8point3.

We also received waivers of our California Flats and Cuyama projects and similarly, given current market indications, we expect to realize considerably higher valuations for those projects.

Secondly, we have been very successful realizing the energy advantage of selling through most of our remaining anticipated supply of our Series 4 product.

Lastly, we've made great strides with the product and commercial readiness of our Series 6 product.

While there's still many challenges ahead, we are pleased with our progress over the first half of 2017, and we'll continue in the future to apply the same disciplined approach, balancing growth, liquidity and profitability that we believe provides our shareholders the greatest long-term value.

Turning to Slide 4. I'll discuss our bookings since our last earnings call.

In total, we have booked 1.5 gigawatts DC in the past 3 months, bringing our year-to-date bookings to approximately 2.1 gigawatts DC.

After deducting the year-to-date shipments through June of approximately 1.2 gigawatts DC, our remaining expected shipments are 3.7 gigawatts DC.

Our 1.5 gigawatts of bookings were geographically diversified in the quarter with the strongest results in the U.S., India, Asia Pacific and Europe.

We also booked new volumes in both Latin America and Africa.

The strong quarterly bookings and increase in our mid- to late-stage bookings opportunity, which I will highlight in my comments later, result to several factors.

First and foremost, the increasing affordability of solar continues to be a fundamental driver of global demand.

We continue to see substantial demand in established markets and the emergence of new markets with significant growth potential.

In order to capitalize on the growth of the global solar market, we began investing in international sales teams several years ago.

We have also entered into strategic partnerships, which allow us to serve markets where we do not have a physical presence.

The bookings for the quarter as a result of these multiyear strategies to focus on sustainable markets and develop long-term relationships with key customers.

Certain developments in domestic and international markets, including strong demand in China, especially for Tier 1 and high-efficiency products as well as a Section 201 case in the U.S. have created an acceleration of procurement timing among customers.

While neither serve to change the underlying fundamental demand in the global market, both have impacted near-term module availability.

Both have also served a firm module pricing in the near term.

However, in the long term, and for as long as the global module supply/demand imbalance exists, we expect the global market demand to remain aggressive.

For this reason, we remain focused on executing the successful transition to Series 6, which is expected to provide us with the most competitively differentiated product.

Relative to the regional highlights for the recent bookings, the United States was our strongest market for bookings with new module volume of over 1 gigawatt DC.

Demand was highly diversified, with projects in California, the better Southwest and the Southeast.

The largest of the bookings was an agreement to supply modules to the 328-megawatt DC Mount Signal project in California.

Demand in India was strong, and in Q2, we booked over 350 megawatts DC of module supply agreement.

The booking's strength was a result of both our strong energy advantage in this region and key customer relationships that we have developed over the past several years.

By working with established customers, we have experienced -- who had experienced the benefits of our modules bring in India, we're able to maximize the realized value of additional energy that our modules provide.

Our bookings in Asia Pacific were diversified across more than 7 countries, highlighted by an additional 41-megawatt DC of project bookings in Japan, the majority of which we utilized Series 6 module.

This brings our contracted portfolio of systems projects in Japan to over 220 megawatt DC, with an even larger number of potential bookings opportunities.

In China, we booked approximately 60 megawatts DC of module supply agreement since the beginning of the year, which represents our first substantial volumes supplied to this market.

In Europe, our bookings this past quarter were primarily in Turkey as a result of our partnership agreement with Zorlu.

In France, we continue to secure additional module supply agreements under the most recent tender process.

Now turning to Slide 5. I'll provide an update on our remaining Series 4 supply.

Our anticipated Series 4 supply across 2017 and 2018 remains between 3.6 and 3.8 gigawatts, depending on when we see Series 4 production at our Ohio plant.

With 1.2 gigawatts DC of volume shipped through the end of June and combined project of module bookings of 2.1 gigawatts DC, we have a remaining Series 4 supply between 300 and 500 megawatts DC.

Entering the year, our goal was to sell the majority of our remaining Series 4 supply by midyear.

With our recent bookings momentum, we have largely been able to achieve that objective.

With our transition to Series 6, we see the sell-through of other remaining Series 4 supply is positive.

However, we are also aware of the significant customer demand remaining in the market and our focus on meeting our customers' needs.

For that reason, we are evaluating options that would provide flexibility to extend Series 4 production beyond the current anticipated shutdown schedule while not impacting our planned rollout of Series 6.

We're still in the evaluation stage, and if any decisions are made, which will be driven by visibility to secure and contract a demand, updated plans will be communicated at the appropriate time.

In addition to strong bookings in the quarter, we've also seen a significant increase in our mid- to late-stage booking opportunities as shown on Slide 6. Now with this metric includes all of our advanced opportunities with shipment dates that extend over the next several years.

Since our first quarter earnings call, bookings opportunities have grown to 8 gigawatts DC, a substantial increase of 5 gigawatts.

The increase in opportunity is a meaningful accomplishment given the strong Q2 bookings of 1.5 gigawatts, the vast majority which were included in the Q1 mid- to late-stage opportunities of 3 gigawatts.

The increase of potential bookings highlight a couple of important trends.

Firstly, we continue to see strong demand of our Series 4 product, which gives us confident in our ability to sell through our remaining supply.

Secondly, this points to the fact that our efforts to prepare the market for our Series 6 product launch is beginning to bear fruit as a substantial portion of the total opportunities presented are for Series 6 module deliveries.

During our last 2 earnings call, we focused on -- in detail on ongoing efforts to -- work with ecosystem partners and customers to ensure a seamless acceptance of Series 6. While these efforts are ongoing, this is an indication that we're beginning to see the impact of this engagement.

It's also important to recognize that 1.9 gigawatts of these mid- to late-stage projects are potential systems project booking.

Year-to-date, the vast majority of our bookings have been third-party module sales, utilizing our Series 4 product.

But this was expected due to the focus on selling our Series 4 by midyear and more importantly, the relatively longer selling cycle of development opportunity.

The higher third-party module bookings does not mean there is any change in our view around project -- future project development opportunity.

If you take our contracted systems pipeline with CODs for 2018 into 2020 time frame, plus the 1.9 gigawatts of potential bookings mentioned, this comes to a potential combined systems business annual average of approximately 950 megawatts.

This estimate does not take into account any opportunities from a much larger earlier-stage development project opportunities, which are not included in our view on Slide 6 but could be converted into bookings.

We are actively continuing to evaluate -- we're also actively continuing to evaluate opportunities globally to acquire development assets or pipelines.

Putting these pieces together and consistent with our previous long-term indications, we continue to have visibility to maintain the systems business of approximately 1 gigawatt per year.

The underlying demand drivers for project development opportunities in the United States and selected international markets such as Japan and Australia are very encouraging.

For example, in the U.S., over the past year alone, we have observed a greater than 50% increase in utility-stated plans to deploy Solar as part of their target resource mix.

We estimate that over the next 4 years, this represents a greater than 10 gigawatt opportunity for Solar.

It's important to note that this fundamentally driven by economics and not solely by renewable or solar-specific mandate.

A decision to install solar has moved from a political mandate to an economic decision.

Additionally, further acceptance of solar is being enabled by utilities' increasing awareness of the capabilities of the utility scale to solar to contribute to grid stability and reliability.

We feel we are well positioned with utility customers and to be a trusted partner as they add increasing amount to solar resources in the upcoming years.

Likewise, corporate customers are increasingly an important source of utility scale to solar demand as they look for ways to hedge their energy cost and improve their demand.

Our multi-gigawatt portfolio of development and module supply opportunities with corporate customers continues to grow, and notably, we now have over 800 megawatts of projects where we have been shortlisted.

With both our Switch Station and California Flats project, we have already demonstrated the ability to meet the needs for large-scale offsite renewables.

Our financial strength and ability to provide integrated solutions, enhanced by our Series 6 module uniquely positions us to meet the needs of this growing market segment.

Continuing onto Slide 7. I'll provide an update on our Series 6 transition.

We are encouraged with our progress thus far in meeting the key milestones we have provided at the beginning of this year.

Through the first half of this year, we are on track and in some cases, ahead of targets provided.

Since our last update, we reached a major milestone with the arrival of our first Series 6 equipment at our Ohio factory.

As shown on Slide 8, our first vapor transport deposition coater has arrived on site and is in the process of being installed.

Additional equipment is being validated at our supplier sites and continues to arrive in Perrysburg as scheduled.

We are steadily moving forward in assembling the front-end of the line, which is expected to be completed in the fourth quarter.

We are also targeting our first complete module off the production line by the end of 2017 or early 2018.

In regard to our Series 6 factory in Malaysia as we mentioned during last quarter's call, we stopped production on 8 lines of Series 4 manufacturing at the beginning of April.

All major production tools have been ordered for this location, and we expect the first tool to arrive in the fourth quarter of this year.

In addition to our efforts in Ohio and Malaysia, we are preparing our existing but never equipped Vietnam factory for Series 6 production.

Moving to Vietnam versus Malaysia, where our third Series 6 factory provides 2 benefits.

Firstly, it avoids ramping down production and therefore, allows us to potentially continue production to Series 4.

Secondly, the Vietnam building is a copy exact of the KLM 5 6 building, which will facilitate in accelerating cost-effective implementation.

While we have not provided the time line for the Vietnam ramp, the key milestones for this location will follow approximately 1 quarter behind those of the Malaysia site.

The Vietnam factory will have 1.1 gigawatts of production capacity once fully ramped.

I'll now turn the call over to Alex who will now provide more detail on our second quarter financial results and discuss updated guidance.

Thanks, Mark.

Starting on Slide 10, I'll begin with second quarter operational highlights of our Series 4 products.

As planned module production was lower in the second quarter at 513 megawatts DC, a decrease of 28% from prior quarter and 35% from the same period last year as we ramped down Series 4 production in our Malaysia facility in preparation of Series 6.

Capacity utilization, which excludes the lines taken out of service, remained high and increased slightly to 99% in Q2 versus 98% in the prior quarter.

Our full fleet conversion efficiency improved to 16.9% in the second quarter, a 20 basis point increase from the prior quarter and an impressive 70 basis points improvement versus Q2 of 2016.

Module conversion efficiency and our best line average 17% in Q2 to 10 basis point sequentially improvement and a 60 basis point improvement year-over-year.

Going forward, we expect the Series 4 fleet average efficiency to level off near our current 17% efficiency as future technology improvements and investment are focused on Series 6. However, keep in mind that the current 17% efficiency level is not indicative of Series 6 efficiency, which will be higher at the time of product launch and is expected to continue to see improved thereafter as we executed on our technology road map.

As a reminder, our record sale efficiency stands at 22.1%.

In the past, we've demonstrated a consistent ability to translate record sale and record module efficiencies into production or regular cadence.

And we expect this process continue with Series 6.

Nameplate efficiency also does not tell the entire story and the energy yield advantage of our modules and the source should be kept in line.

Whilst our focus remains on realizing the potential of Series 6, our Series 4 product remains very competitive from both an efficiency and energy advantaged standpoint.

Turning to Slide 11, I'll discuss some of the income statement highlights for Q2.

Note that I'll be discussing some non-GAAP measures such as adjusted operating expenses, adjusted operating income and non-GAAP earnings per share.

And please refer to the appendix of the presentation for the accompanying GAAP to non-GAAP reconciliations.

Net sales in the second quarter was $623 million, a decrease of $269 million compared to the prior quarter.

The decrease in net sales resulted primarily from lower systems project sales, partially offset by higher third-party module volume.

The reduction in systems project sales is primarily a result of recognizing 100% of the Moapa project in Q1.

In the second quarter, we sold Switch Station to EDF Renewable Energy and recognized the last percentage of the revenue.

We will recognize revenue on a percentage of completion basis until the project reaches COD, expected in Q3 of this year.

Switch Station is an example of how First Solar can work jointly with utilities and commercial customers to meet their needs for clean, affordable solar power.

This is the first project we sold to EDF in renewable energy and we look forward to future partnership opportunities.

Also with the sale of this project, we now recognize another important component of our guidance for 2017.

As a percentage of total quarterly net sales, our solar power systems revenue, which includes both our EPC revenue and solar modules used in systems projects decreased to 63% from 92% in Q1.

Third-party module sales were $228 million in Q2 versus $71 million in the prior quarter.

Note that third-party module volume recognized in the quarter was meaningfully lower than third-party shipments based on timing and meeting our revenue-recognition criteria.

Gross margin improved 18% in the second quarter from 9% in Q1.

The increase in gross margin resulted primarily from the different mix of systems projects recognized between the quarters.

The lower Q1 gross margin was related to the sale of our Moapa projects, which as we've indicated in the past had a gross margin profile lower than we would normally expect from the system sale.

The sale of Switch Station in Q2 is more representative of our targeted gross margin range on a system project sale.

Gross margin in our components segment was 17% in Q2 versus 26% in the prior quarter.

Operating expenses, excluding restructuring and asset impairment charges, were $79 million in the second quarter versus Q1 adjusted OpEx of $72 million.

The sequential increase in operating expenses is primarily due to the higher plant start-up costs associated with Series 6. We expect plant start-up cost to continue to increase in the second half of the year as Series 6 work in Ohio and Malaysia intensifies.

Highlighting the impact of the OpEx reduction initiatives we've undertaken, our OpEx, excluding restructuring and plant start-up, has decreased by 27% versus the same period in 2016.

Restructuring and asset impairment charges to accelerate our Series 6 transition by $18 million in Q2, a decrease of $2 million from the prior quarter.

The Q2 charge is primarily related to net losses on the disposition of Series 4 and Series 5 manufacturing equipment.

Excluding restructuring-related items, adjusted operating income in the second quarter was $32 million compared to $12 million in Q1.

The increase is due to higher gross margin on Q2 sales, partially offset by the increase in production start-up expense.

On a GAAP basis, our operating income for the quarter was $14 million.

We had a tax benefit of $40 million in the second quarter compared to $6 million of tax expense in Q1.

In Q2, we recognized a $42 million discrete tax benefit resulting from the acceptance of our election to change the tax status of a foreign subsidiary.

As we indicated on our last earnings call, this amount could have been up to $55 million and was not reflected in our guidance at the time.

Note also that both quarters have tax benefits associated with restructuring charges.

Second quarter EPS was $0.50 on a GAAP basis and excluding restructuring and asset impairment charges, $0.64 on a non-GAAP basis.

This compares to Q1 GAAP EPS of $0.09 and non-GAAP EPS of $0.25.

Relative to our indication on the prior earnings call, Q2 non-GAAP EPS was higher than anticipated due to both the $42 million tax benefit and the initial recognition of the Switch Station projects.

I'll next discuss select balance sheet items and summary cash flow information on Slide 12.

Our cash and marketable securities balance ended the second quarter at $2.2 billion, a decrease of approximately $217 million from the prior quarter.

Our net cash position declined by $261 million to $1.9 billion.

The decrease in our cash balance is primarily timing-related as we'll receive payments for the majority of the Switch Station's project sale in Q3.

Q2 net working capital, which includes the change in noncurrent project assets and excludes cash and marketable securities, increased by $171 million.

The change is primarily due to an increase in accounts receivable from higher module sales in the quarter and the timing of payment from the Switch project sales.

Total debt at the end of Q2 was $321 million, an increase of $44 million from the prior quarter.

The increase resulted primarily from issuing project level debt for projects in Japan and Australia.

Cash flow used in operations were $166 million in Q2 versus cash flows from operations of $493 million in the first quarter.

Free cash flow in Q2 was a negative $271 million compared to positive free cash flow of $380 million last quarter.

Capital expenditures were $105 million compared to $113 million in the prior quarter.

Moving on to Slide 13, I'll review our updated full year 2017 guidance.

As it relates to project sales in our guidance, our expectation of selling both the California Flats and Cuyama project in the second half the year remains unchanged.

Both projects will offer to 8point3 earlier this year but has since received the waiver of the negotiation period from 8point3, providing First Solar the right to offer and sell this project outside the YieldCo in accordance with the terms of the ROFO agreement.

In the case of Cuyama, we've recently entered into a sale agreement for the project, which we expect will close in the third quarter.

The sale of California Flats is progressing well and we're in late-stage negotiations with the buyer.

In terms of international projects, our guidance also assumes we sell a small number of our Japan projects and some India projects in the second half of the year.

As we progress further into the sales process on Switch Station, California Flats, Cuyama and certain of our international projects, our expectations for the sale value of these projects have increased.

This additional information, plus an increase in expected volume of third-party module sales, we'll have to increase our net sales guidance range by $150 million to a revised range of $3 billion to $3.1 billion.

We're raising the midpoint of our gross margin guidance 400 basis points to a revised range of 17% to 18% as a result of the expected increase in the sale of value projects and based on continuing cost improvements.

We revised our GAAP operating expense range to $370 million to $395 million.

The low end of the range has increased as we now have better line of sight to remaining operating expenses.

Note that total plant start-up expense for 2017 is still anticipated to be around $50 million.

We've lowered the top end of the operating expense guidance range by $10 million as a result of our reduction in our expected restructuring charges.

The revised range of restructuring-related items is now $40 million to $55 million with approximately $38 million of charges incurred through the first 6 months in the year.

The low end of our non-GAAP operating expense range has also been raised by $10 million, resulting from higher variable compensation and higher project-related transaction structuring and sale costs, which have driven higher project values.

Our non-GAAP operating expense excludes the impact of restructuring and asset impairment charges.

We are raising our EPS guidance by $1.75 as a result of better economic value expected from project sales and improved operational performance.

In addition, we're raising EPS as a result of the $42 million discrete tax benefit in Q2.

Our revised non-GAAP EPS range is now $2 to $2.50.

Non-GAAP EPS range has been revised between $1.55 to $2.20.

Note that our non-GAAP EPS assumes a full year tax benefit of between $25 million and $30 million.

As it relates to the second half of the year, the quarterly earnings profile will be highly dependent on the timing of the sale of California Flats and various international projects.

Depending on the project close timing, particularly for California Flats, second half earnings could be much more heavily weighted to Q3 versus Q4.

The fourth quarter result is expected to have relatively higher plant start-up expense and lower module shipments.

In terms of cash, we've raised our ending 2017 net cash guidance range to $2.1 billion to $2.3 billion, a $600 million increase versus prior guidance.

Expected higher net cash balance as a result of several factors.

Firstly, approximately $125 million of the increase is attributable to the higher expected value of project sales.

Next, an additional $125 million as a result of our lowered 2017 CapEx spend.

Keep in mind, the CapEx reduction is timing related, and we expect to spend that capital in 2018.

Additionally, we expect project development spend in 2017 to now be around $200 million lower primarily due to timing.

Some cases for our updated plans use Series 6 modules.

We will now complete more development work on those projects in 2018.

The remaining updates on that cash guidance is a result of improved working capital, primarily associated with our module business.

Operating cash flow guidance has likewise been increased as a result of the expected higher project sale pricing, lower development spend and improved working capital.

This expected working capital benefit is attributable in part to our plan to recycle capital faster from our systems business.

It's also a reflection of the decreasing capital intensity of our project business as cost continue to decline.

Overall, we're very pleased with the cash generation the business during this transition period.

This provides us with great flexibility to invest in the future of our technology and manufacturing capacity while still having adequate capital for our systems business and other priorities.

I'll now summarize our second quarter 2017 progress on Slide 14.

We had solid financial results in the second quarter, boosted by our sale of our Switch Station projects.

Our net sales were $623 million and non-GAAP EPS was $0.64.

Our ending cash balance is $2.2 billion, with $1.9 billion of net cash.

We raised the midpoint of our GAAP EPS by more than $1.80, our non-GAAP EPS by $1.75.

Our Series 4 module efficiency remained solid with a full fleet average of 16.9% and lead line efficiency of 17% in Q2.

Our bookings for the quarter were impressive with 1.5 gigawatts of new business contracted and mid- to late-stage opportunities of over 8 gigawatts.

Our Series 6 transition is progressing well with the arrival of tools at our Ohio factory and installation is underway.

And with an increase of Series 6 mid to late-stage opportunities, we're pleased with the initial response from our customers around Series 6.

And with that, we conclude our prepared remarks, and we'll open the call to questions.

Operator?

(Operator Instructions) And we'll go to Philip Shen.

The first set of questions I have are around capacity.

You mentioned in your prepared remarks that you're evaluating maintaining Series 4 capacity beyond the gigawatt that you have talked about.

Can you give us a sense of how much that might be?

And then as it relates to Series 6, and once you get Terra 0, 1, 2 and 3 up and running, I believe you'll be at 4 gigawatts of capacity.

When do you think you could hit that 4?

Can we get there in 2019?

What is your base case expectation for 2019?

Yes.

So I'll take the first one on the Series 4 capacity and I'll let Alex talk to Series 6 and what the first 3 Terra plant capacity was provided to us, and the time line of expected production in 2019.

As it relates to Series 4, one of the things we said in call, I want to make sure people understood as well is that we've now made a decision, instead of doing the third Series 6 plants in Malaysia, we're actually moving that now to Vietnam.

So as you all know, I think, or most of you know, we have a building in Vietnam that we never began production in, but we will now move Series 6 third factory to Vietnam.

And that provides capability then to look at our KLM Plant 3 and 4 as it relates to the timing of how long we'll continue to run that production.

So we have optionality in really the first 4 plants in KLM, KLM 1, 2, KLM 3, 4 as it relates to continuing Series 4 production into 2018 or through all of 2018.

Current plan assumes that all that production would be stopped in the middle of '18.

We also have capability in 2 lines in Perrysburg.

We had 6 lines in Perrysburg.

We've actually shut down production of the south building which maintained 4 production lines.

We've had that production, obviously, ramped that down so that we can ramp up Series 6, but we still have the other 2 lines in Perrysburg that we can evaluate.

So we have -- call it, 10 lines as it relates to decisions that can be made as it relates to Series 4 and how long we continue that production.

We're evaluating alternatives right now.

There's a lot that's going into our decision-making as it relates to running Series 4 production.

Part of it is underlying project economics, module economics.

Making sure we can capture what we think is fair value for that product.

The other is, obviously, most important is making sure none of them has any impact to distracting or delaying the rollout of Series 6. If there's anything that would cause a delay in Series 6, we would not do that.

We're 100% committed to getting the most competitively differentiated product into the market as fast as possible.

So we don't want to do anything with Series 4 that would jeopardize our Series 6 launch.

The other thing I would say is that as it relates to Series 4, I want to capture as much of the value stream as possible.

So I'm not just looking to capture module sales per se, if I could find a way to continue to run Series 4, capture all the way through development, or through EPC, through OEM, capture more of the value chain, those are the things that will inform our decisions around what we do with.

We feel very fortunate to have the optionality around Series 4. But again, the top priority is we won't do anything to jeopardize the launch of Series 6. So I'll pass the Series 6 question over to Alex.

Yes, as it relates to Series 6, the plan we've laid out, which is about $1 billion of capital spend, has us with the first line in Perrysburg at 550 megawatts and then a full high volume manufacturing line about 1.1 gigawatts.

So we will go first to KLM and then, as Mark said, over to Vietnam for the second Terra factory.

If you look at that spend now, that gets us to about 3.5 gigawatts of production in 2019 and has a leading exiting 2019 just under 4 gigawatts of capacity.

We'll go to Krish Sankar, Bank of America Merrill Lynch.

This is Chirag Odhav on for Krish.

Could you just comment a little bit on any changes in customer activity, resulting from the ongoing Suniva trade case and have you noticed an increase in customers looking to secure longer-term pricing as a result?

Yes, so there's 2 sides on the customer activity.

One is fundamental demand for solar and for our power plants.

I don't -- obviously, tier 1 hasn't changed.

That demand profile at all and what we said in our prepared remarks is that there is a tremendous increase in demand for solar, driven by utilities and looking at long-term integrated resource plans and understanding the affordability and the competitiveness of solar, so that's, obviously, driving more demand.

We're seeing a global increase in solar demand, again, it's driven off of demand elasticity so as the prices of solar come down, we're seeing a significant increase in demand associated with that.

We're also starting to see, as we said in our prepared remarks, a pretty significant increase in utility scale, commercial and industrial opportunities, and we have pretty robust pipeline from that standpoint.

So the fundamentals as it relates to solar hasn't been impacted or hasn't been -- has not at all driven by anything that may happen have on a trade case.

The trade case here in the U.S., is it driving potentially customers to make their procurement decision sooner than they would otherwise?

I think there's probably some of that.

I think people are trying to get access to supply in advance of any ruling that may be made, trying to protect their projects as it relates to their embedded economics and insecure a module supply, so I do think that, that is happening in the marketplace.

And depending on how the trade case plays out, we may see more of that activity here over the next quarter or so.

We'll next go to Tyler Frank, Robert Baird.

Can you just talk about how sustainable you think current economics are?

You mentioned both on the project side, seeing better value for the projects that 8point3 give you a waiver for.

And then obviously, on the module side, it seems like prices have firmed up.

So as we look out 2 to 4 quarters from now, what are your expectations for overall the demand and supply balance and how should we think about the long-term sustainability of the current pricing profile?

So I'll take the discussion on module and Alex will take more of the discussion on the system side.

Look, I think the reality is that the module prices, and we'll use the U.S. as an example, are such -- relatively insignificant component of the overall levelized cost of energy.

As we said before, cost of capital is even a more significant impact, right?

But the balance of system cost now are getting to be more impactful to some extent if you look at the impact in burgers and trackers, you know you're seeing a more significant component of the overall LCOE.

If you look at the module as an example, it will vary upon regions that you're in, but the fact it will be somewhere in the range, call it, around $0.03 of module price drives about $1 of PPA price.

So if you're thinking -- if you're pricing at $30, and you see a module price move by $0.03, that adds about $1 to the PPA price, that's not going to change fundamental demand associated with procurement decisions, right?

$1 PPA price isn't going to change somebody's decision around whether they're going to move forward and procure.

So I think that -- look, it's a competitive market, there is, obviously, as we said before, there is excess supply and oversupply in the industry.

It is a little bit tighter on the higher efficiency, higher quality products.

So that you have to bifurcate the market a little bit that way.

But as I look over the next few quarters, yes, things will continue to be competitive, but I think we're relatively tight range of module price.

And the other thing I would say is that what we saw some of the more aggressive pricing behavior from some of our Chinese competitors in particular, I think there are some signs of fatigue.

I think some of them made statements that they've reduced shipment guidance for the year because they're going to focus on profitability.

So look, this industry moves very quickly, so to try to make a definitive call for something 2 to 3 quarters out is never easy.

But I would say at least for the second half of this year, we'll probably in a relatively tight range relative to where module prices will move.

Hey, Tyler, on the systems side, I'll say that the outlook is, I think fairly sustainable.

For us, there are already 2 things that work here.

One is we are seeing a lot of interest in U.S. asset ownership at a time when there are relatively few large-scale quality assets available in 2017 and into 2018.

So despite some potential headwinds around tax reform and interest rates, we're not seeing a significant change in cost to capital at the moment, and we're seeing a lot of money in the infrastructure space and the post equity space looking to go to work.

So we're seeing a lot of interest there.

I think the second thing for us and this relates a little bit to 8point3 is the competitiveness of the YieldCo and it specifically relates to the assets that we have this year and the ROFO process and the value associated with those, and California Flats is a big driver of our guidance change right now.

When we started marketing that asset, we were aiming to structure a deal for -- to keep our residual interest for 8point3.

As we progressed through that, it became clear that the structure in itself was creating a loss of value and at the same time, the yield profile of 8point3 was changing.

So when you combine those 2, it meant that 8point3 was clearly not the best buyer for that asset.

And we took that back out to the market, and we found a simpler and less structural deal with a much higher value that influence not only California Flats, but how we think about other assets in the future.

So when I look at those two, I think, partly, there's just a very robust market for U.S. asset sales today and the amount of capital and the cost of capital available is very good.

And secondly, the YieldCo is not really a viable buyer today if the new U.S. utility scale assets are currently capitalized.

Vishal Shah with Deutsche Bank.

Just have a question on the gross margin outlook.

I know you guys have previously talked about how margins could be potentially under pressure until Series 6 is up and running in the second half.

Do you -- with the current margin pricing environment, what's your current outlook on component market margins?

And then for Series 6, are you still looking at about 1 gigawatt of 2018 shipments?

And then finally, as far as your capacity allocation is concerned for projects that you intend to complete and sell in 2018, can you maybe talk a little bit about what that number would look like.

This year, for example, I think you have about 0.5 gigawatt of capacity that's being allocated to internal projects.

How would that number change in 2018?

Vishal, I'll take the gross margin outlook and talk a little about the project side of '18.

And Mark can talk on the Series 6 '18 shipment.

On the gross margin, so what we've said before and remain true is that as we stop, looking to improve the Series 4 module and focus on the Series 6 module, that, that module is going to be most challenged just before we ramp it down.

So right now, if you look at our gross margin, you can see that this quarter is a little lower than we've seen in past quarters.

Part of that is due to ASP declines from volumes that were booked in previous quarters, and then there's a small impact from underutilization cost as well.

So as we've taken down our KLM 5 and 6 line, which was one of our lowest cost locations, the blended cost when you include our Perrysburg factory being a larger component of the total, Series 4 increases slightly.

So I'd say that on the module side, I think the margins you're seeing today are probably relatively sustainable over the next few quarters, and we've seen pricing firm up in the U.S. a little bit.

On a project side, we haven't guided to 2018 so we don't have number to give you today, but if you look through the Q, it will come out later.

It will give you a view to what's in the pipeline and the COD dates can give you a rough guide to what we look at in 2018.

Relative to Series 6, we still are on target for about 1 gigawatt of production for next year.

So that lines up to -- as we said everything's on schedule and progressing as anticipated.

So as we think about '18 and still what our expectations are.

And again, the constraint isn't necessarily our own ability to ramp, it's actually getting the toolsets from our vendors and obviously, there's a relatively long lead time from that perspective.

The other thing I would say around the capacity allocation systems while we haven't guided, as Alex indicated, but we did indicate in the call and I think it's the right way to look at it is that when you look at our contracted pipeline, which you'll be able to see more around that in the Q when it's filed, plus our late-stage opportunities that we're pursuing right now, call it 1.9 gigawatts, we are lined up very well for '18, '19 and '20 on an average basis.

It will be approximately 1 gigawatt of systems business.

So I think that's the right way to look at it.

We will continue to optimize the timing of the shipments and the recognition around those assets to optimize value creation from that perspective.

But I do I feel comfortable where we are right now with maintaining that annual volume around 1 gigawatt for our systems business.

Brian Lee, Goldman Sachs.

I have a couple of them.

Maybe first off.

Mark, can you comment on the pricing environment for modules, just in the U.S. but also globally?

I know there is a bit of a difference there.

And have your raised prices in the U.S. specifically and by how much, if you can comment?

And then just maybe lastly, on that point, how are you pricing bookings especially for deliveries in 2018 given some on the trade case uncertainty?

The followup I'd have is just around the systems visibility for the raised guidance.

Can you quantify how much of a margin improvement you are seeing from the structuring change you mentioned, Alex?

And then, where do you think margins on systems are better than modules in the second half of the year given some of the price bifurcation we're seeing in the market?

All right, I'll take the first two and I'll have the last one.

As it relates to the pricing environment, talk to the U.S., just clearly things have firmed up, right?

I said that in my prepared remarks as well, so the pricing has firmed up in the U.S. Again, there's a tremendous amount of demand right now across all segments of the market, which is firming up that pricing.

Internationally, it depends on the market.

Some I would argue are stabilizing and firming up a little bit.

Others, I would look -- would argue are still very aggressive.

So it's hard to kind of give you a generic statement relative to the international side of the house.

But on average, I would say the global, if you take the U.S. combined with international markets it's relatively -- the pricing has firmed up over the first half of this year relative to where we exited the end of last year.

The -- in terms of how we're thinking about the trade case and how we're selling forward, we continue to try to best position our pricing in the marketplace relative to the competition and sell through the initiation(inaudible) and value creation that we have inherent in our module, which is the energy yield advantage.

As it relates to the trade case, it's still -- an uncertainty at this point in time, as it relates to, a, that's going to happen, b, to what extent there would be an impact.

So we haven't really informed our pricing decision per se as it relates to if something were to happen, I think as we get closer and you see better line of sight, we'll evaluate that.

One thing I will say, though, is our general approach especially as we position Series 6 into the marketplace is we will capture fair value for the technology that we provide to our customers.

But I'm not looking at this as some opportunistic ASP grab that we could get into the marketplace.

I mean we're going to engage customers from a relationship standpoint and long-term partnership perspective and capture the right appropriate value for the product, not necessarily trying to be overly optimistic because of the potential trades that either may happen or may not happen.

Brian, to your second question.

If you look at the EPS guidance, we've raised that $1.75.

There is a tax benefit in there, about $42 million.

The rest is associated mostly with the systems business.

On Cal Flats, there are 2 things that work.

One and we brought this up, I think, on a previous call is though, is a specific development related risk item that was not in our guidance before, which has now been resolved, relative to property tax, so that provides significant uptick to value and then the remainder of that then comes from the structuring and taking the deal out of the YieldCo and selling to a third party.

But from a value perspective, I would look at the EPS raised minus the tax mostly attributable to the systems business.

I think what Alex did say, I think Brian was asking what does the second half margin profile look like between systems and modules.

I mean, one of the things that Alex indicated in his comments as it relates to -- as we ramp down production of Series 4, again, there is under absorbed, underutilized cost structure there that will create a little bit of a headwind on our Series 4 cost, all right?

So that by default then, what we would potentially expect is to have lower gross margin which we saw in Q2 versus Q1 because of the impact.

And if you use a penny as an example.

If there's a penny underabsorbed cost that now has to be weighted down towards Series 4, you're talking something that's going to be 3% to 4% gross margin, so a small delta could have a significant impact to gross margin percent.

And we saw a little bit of that here in the second quarter.

Colin Rusch, Oppenheimer & Co.

Can you talk a little bit about the impact of lower energy storage prices on demand overall for projects at this point and how you're developing going forward in lieu of what we're seeing in terms of that cost decline on energy storage?

Yes, so it's -- especially in certain markets, the impact of storage -- and it's really it's interesting because not only in mature markets maybe have a more fundamental issue that they're trying to deal with right now and everyone generally referred to as the Dr. Phenomenon.

It's not only there, whether it's an increased interest on storage and the integration of PV with storage.

We're going to start to see as we see some of the demand now migrate into the Southeast as an example and some of the utilities are early on in their journey for solar and it's now strategic and integrated into the long-term resource plans.

They want to understand storage as part of that long-term solution.

So we're seeing it on both sides.

We're more mature markets but even more emerging markets as well.

The other thing that is going to continue to concern is around great reliability and stability when you see increased renewal penetration, and we demonstrated through our work with Cal Flats and (inaudible) and reports that we've issued and validated by others is that utility scale of solar, obviously, can improve, it can provide services that actually improve reliability and stability of the grid.

And that's another item that a lot of utilities are asking a lot more about and trying to have a better understanding whether their proper power plant control that enable that.

And we feel very good about our offer in that regard.

It's clearly going to be another inflection point for demand generation.

It enables a much higher solar penetration in a number of key markets and what's happened with the battery side of the equation, the cost have come down significantly, and you're seeing very compelling PV plus storage, and we're actively engaging and participating in a number of RFPs.

In some cases, even more almost bilateral negotiations with a couple of customers to make sure that we can demonstrate our innovation and thought leadership in this area.

But I do think it's going to have a significant impact on demand as we look forward in the next 5 to 10 years.

Sophie Karp, Guggenheim Securities.

Can you comment a little bit on your strategy with 8point3 Energy Partners at this point?

Are you still pursuing the sale of the stake and the expectation with the pricing and the project values that you're seeing impacted it in any way?

Yes, I mean, we can't go into a lot of details as it relates to 8point3, but we are still exploring strategic alternatives which would include the sale of our interest, specifically it could include something more than that and on the interest from a market perspective.

I think as Alex indicated, and I said in my prepared remarks as well, is that we've captured tremendous upside by selling our high-quality assets outside of 8point3.

And if you look at the delta where the share price would have to be to effectively give us equivalent economics on an asset-light Switch Station, it would actually have to trade above where the IPO was.

And there's no indication of anything in that range and I don't think we'll ever see the yield that we saw at that point in time coming back anytime soon.

So I think where we stand, we know that we can capture better -- we can get better value capture for our assets with other buyers.

It creates more of a competitive dynamic to the sell down profit.

So I don't think, strategically, anything has changed from our perspective relative to 8point3.

Our strategic valuations are still ongoing and I have no other update beyond that.

Our final question is from Pavel Molchanov, Raymond James.

Let me go back to the tariff issue.

There have been some estimates from the SEIA and others that if the Suniva request were granted, it would cost something like 100,000 jobs in the U.S. solar value chain.

Because you guys are both hardware vendors and project developers, I guess my -- the way I would frame it is, what do you think is the optimal outcome?

Are you rooting for Suniva's request to be granted or not?

So it's interesting.

We're part of SEIA.

We love SEIA, but I can take you back to the first trade cases that have been put in place in 2012, 2014 time frame, same argument was there.

This should be a complete disruption of jobs, all these jobs are being created, we're going to be a tremendous risk, we're going to lose hundreds of thousands in solar job.

If you take that timing and move it forward, nothing more than -- the only thing that happened is the industry has continued to drive.

Nothing fundamentally changed and disrupted the demand profile for solar.

There were no job loss, there's jobs created over that period of time, okay?

So I know the rhetoric there and people like to lead with that.

But I think we got to step back and look at the reality.

And I think if look at most recent history, you would argue that the reality is something much different than maybe what people are saying.

What I would say is I go back to statements we said in our last call.

There is a tremendous oversupply in the industry, right?

We believe in free and fair trade, to the extent there's not fair trade then there's an element of enforcement potentially that needs to be put in place.

We're not a part of the trade case, we're not anticipating to be a part of the trade case.

The process we rolled out right now, there's still in the valuation injury and determination.

That will be made sometime by the mid- to late September.

Once -- if there is a determination of injury at that point in time, they'll move forward into a remedy phase, which will last, again, another 6 to 8 weeks.

We would potentially -- if we do get involve, potentially be involved as it relates to input around remedies.

I do think that modest type -- if there is a determination of injury, a modest type of remedy, will not be harmful at all to the industry and I think we'll continue to thrive and more jobs will be created.

If the current proposed recommendations around minimum price is in (inaudible) go up into the $0.40 tariff range and minimum price of close to $0.70, I think that could have an impact.

Something more modest, let's call it $0.10 to $0.12, maybe $0.15, I already said $0.03 of module price is $1 of PPA price.

So if you have $0.10, you have $3 to a PPA price.

Something going from $30 to $33, I don't think will be disruptive at all to the underlying demand profile for solar.

It won't change the -- it won't create any type of risk to job creations.

And I think, initially, issue will continue to thrive.

We already reference the storage impact so now they're kinking the curve of making solar more competitive, viable and sustainable.

So look, we'll continue to evaluate, we'll continue to provide input if it gets to a remedy standpoint.

But I don't see that there's the -- that the rhetoric that there's going to be some tremendous disruption of job creation in the U.S., I don't think that's going to happen.

And hopefully, what it'll do is enable more manufacturing jobs in the U.S. I think that's another opportunity that this could result in.

And whether we do it in our plant in Ohio or others potentially bring more manufacturing in the U.S. I think that would be a great thing.

Mr. Haymore, that does conclude today's conference call.

We thank you all for your participation, and have a great day.