第一太陽能 (FSLR) 2017 Q3 法說會逐字稿

Good afternoon, everyone, and welcome to First Solar's Third 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 conference over to Steve Haymore for First Solar Investor Relations.

Mr. Haymore, you may begin.

Thank you, Ashley.

Good afternoon, everyone.

Thank you for joining us.

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

A copy of the press release and associated presentation are available on the Investors section of First Solar's website at investor.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 financial results for the quarter 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.

I'll begin by highlighting some of our third quarter financial and operational results.

Firstly, we achieved record quarterly net bookings of 4.5 gigawatt DC, which brings our year-to-date net bookings to 6.7 gigawatts.

The strong bookings demonstrates the tremendous customer acceptance of our Series 6 product and the impact of market factors that are leading to an acceleration of procurement timing by certain customers.

Our financial results for the quarter were also strong, with revenue of over $1 billion and EPS of $1.95, driven by the sale of both our California Flats and Cuyama projects.

Turning to Slide 4. Our Series 6 transition continues to progress according to plan and remains on schedule.

We are now more than halfway through our 18-month Series 6 journey which we started in November of 2016.

I'm very pleased with the impressive results the team has delivered so far.

At our Ohio factory, the front end of our Series 6 line is now almost completely installed, and most of the equipment is operational and in various phases of acceptance testing.

We are also beginning the significant milestone as we run glass through the core semiconductor equipment, including the cad tel core.

At our Malaysian factory yesterday, our first vapor transport deposition coater arrived on site, and the process of installing the tools will ramp up in the coming weeks with the added benefit of the learnings from our Ohio factory.

While we will save further updates until our Analyst Day on December 5, we continue to be excited by our progress.

Before continuing further, I'll briefly comment on the trade case.

As we have said in the past, we are not a party to the proceedings which excludes thin film.

However, we did recently make our voice heard on this matter, but only after several respondent parties opposing import release repeatedly cited First Solar in their briefs and testimony to the ITC.

With our name being used in this way before a U.S. investigative agency, we couldn't stay silent, particularly when we recently were forced to eliminate hundreds of manufacturing and nonmanufacturing jobs in the United States.

As we said, we have consistently witnessed seemingly irrational market behavior from foreign companies who have continued to expand production capacity despite years of low or negative returns on investment.

We believe that an effective and reasonable remedy on crystalline silicon PV imports can indeed coexist with continued growth in the U.S. solar demand in all segments of the solar industry.

We believe that U.S.-based manufacturing is essential to sustainable economic prosperity, and we will support the U.S. government actions that best serve American workers, manufacturers and the solar industry overall.

Continuing on to Slide 5, I'll discuss our bookings in more detail.

We have booked an excess of 5 gigawatts since our last earnings call, and netted against this volume is approximately 500 megawatts previously booked with customers under framework agreements, which have now been terminated.

The net result is quarterly bookings of 4.5 gigawatts DC, which after deducting year-to-date shipments through September of 2.1 gigawatts, brings our total remaining expected model shipments to approximately 7.4 gigawatts.

As a note, the delivery timing of these bookings stretches over the next several years and includes volumes planned for shipment into 2020.

The strength of Q3 bookings is a reflection of the positive customer response to our Series 6 product, as highlighted by the number of Series 6 contracts included in these bookings.

In certain cases, when contracting this volume, we have included flexibility which allows us to meet our customers' demand with either Series 4 or Series 6 modules.

Approximately half of the Q3 bookings have this contractual flexibility.

Prior to this quarter, the only Series 6 bookings were designated for our own captive projects, but with these recent contracts, we now have a substantial Series 6 pipeline with third-party customers.

As mentioned last quarter, there has been a confluence of events that has driven a recent strong bookings performance.

Firstly, global demand continues to be strong as solar becomes increasingly economical relative to other sources of generation.

Secondly, while not impacting the underlying demand fundamentals, we have seen the Section 201 trade case in the United States accelerate module procurement timing by some customers.

Lastly, the recent surge of demand in China has created, in the near term, a relatively tight supply of Tier 1 module manufacturers.

Looking to the future, we anticipate the historical global module supply-demand imbalance to persist and, therefore, expect the global pricing environment for modules to remain aggressive.

For this reason, a successful transition to Series 6 remains a top priority, as it's expected to provide us with the most competitively differentiated product and best position the company for long-term profitable growth.

I wanted to highlight an important point on Q3 bookings.

While most of the bookings for the quarter were third-party module sales, we expect to eventually add additional scope, such as EPC and O&M services to some of these arrangements, which would increase our overall systems bookings number.

Historically, a developer with contracts for EPC and O&M services concurrently was selecting a module provider.

However, in the current U.S. market environment, the procurement process is, in some cases, inverted, with developers looking to secure module supply in advance of contracting for additional services.

For example, our bookings in the U.S. includes approximately 85 megawatt DC of systems bookings for a project in Florida.

The module bookings with this customer are much higher than 85 megawatts, and we are in the process of negotiating EPC agreements for the remaining volume.

However, until the EPC agreements are signed, the bookings will be reflected as module-only sales and not shown in our systems project pipeline.

Once any additional agreements are signed, while it will not be reported as incremental megawatt bookings, it will represent revenue and margin on top of the original module bookings.

We are excited to have been awarded this volume with a strategic customer in Florida.

Most importantly, this volume is representative of growing utility-owned generation market segment as our customer plans to own and rate base these projects.

Another important in growing market segment is the utility-scale of commercial and industrial segment.

In the U.S., we signed a greater than 400 megawatt DC supply agreement for a project that will further a major corporation's goal to run its operations with 100% renewable energy.

More detail will be available in the future.

Adding to our growing track record of successes with corporate customers, we have recently been awarded a greater than 100-megawatt PPA for a project that will supply another major corporate customer with clean and affordable electricity.

Once we finalize and sign the PPA, this opportunity will be reflected in our bookings.

Both of these projects highlight the growing demand from corporate customers for solar power and the expertise that we can provide to meet their needs.

The remaining bookings in the U.S. were primarily for module sales in the Southeast.

Turning to our international bookings.

In Australia, we signed a new module supply agreement of 240 megawatts.

This brings our total contract to delivery pipeline to over 500 megawatts and highlights our leadership position in supplying large-scale solar projects in Australia.

In Japan, we booked a 21-megawatt DC systems project, which now brings our total contracted pipeline in Japan to over 240 megawatts DC, Illustrating the opportunity for more Japan systems booking in the future, our mid- to late-stage bookings opportunities are greater than 300 megawatts DC.

In other parts of Asia Pacific region, we have also had significant module sales in both Malaysia and China.

With the bookings in the quarter, we have now sold through our approximately 3.6 gigawatts of Series 4 module supply based on our current production plan.

As we indicated in the last earnings call, we have been evaluating options that could extend Series 4 production beyond our current operating plan.

These options could add up to 1 gigawatt of additional Series 4 supply in 2018.

As we said previously, any decision to extend Series 4 production would not impact our previously announced Series 6 rollout plans.

We intend to make a decision on this later this year and will provide an update at the upcoming Analyst Day.

Turning to Slide 6. I'll next highlight our mid- to late-stage bookings opportunities.

Similar to prior quarters, this metric includes all of our advanced-stage opportunities which have shipment dates that extend over the next several years.

Despite the current quarter's strong bookings, our potential bookings opportunities of 6 gigawatts have been very resilient and decreased only 2 gigawatts relative to the prior quarter.

For reference, the opportunities shown here are almost entirely for Series 6 modules.

The growth in our mid to late-stage Series 6 opportunities over the past several quarters demonstrate the extremely positive response from customers to our Series 6 product, both in the United States and internationally.

Included in the Series 6 -- excuse me, included in the 6 gigawatts of potential bookings are over 2 gigawatts of systems opportunities.

This includes U.S. utility RFP opportunities, over 800 megawatts of projects with corporate customers and international opportunities in Japan and Australia.

A meaningful portion of this volume has already been awarded and is in contract negotiations.

If eventually booked, shipments to these systems projects would fall primarily in 2019 and 2020.

Also, keep in mind that we have a much larger number of early-stage systems opportunities not included in this total.

The number of mid- to late-stage opportunities gives us confidence in achieving our target of an average of 1 gigawatt per year of systems business.

Lastly, turning to Slide 7. I'll highlight some key points from our recently published 2017 sustainability report, which is available on our website.

At First Solar, environmental and social responsibility are a part of our core values.

These are values we embrace each and every day as we produce modules and construct power plants that provide clean, affordable energy.

To illustrate this point, with over 17 gigawatts of modules installed worldwide, our PV solutions displaced more than 12 million metric tons of CO2 per year, which is equivalent to powering more than 8 million homes annually based on worldwide averages.

We are proud of the positive environmental impact our technology has on society and the progress we continue to make.

While solar technologies have a lower impact on the environment than fossil fuels generation, a distinguishing feature of our technology is the substantial benefit it possesses compared to other PV modules.

On a lifecycle basis, our thin-film modules have the smallest carbon footprint, lowest water use and fastest energy payback time in the industry.

In fact, a recent third-party study evaluated the environmental footprint of 5 different PV technologies and found that the impact of First Solar's PV systems are about 2/3 lower than the average PV system.

Our lower carbon solar technology not only has positive environmental benefits, but also provides a competitive advantage in commercial discussions.

For example, as we announced early this year, we were awarded 107-megawatt module supply agreement by Photosol as part of the third round of the French tender.

Photosol's decision to select First Solar's module was not only a result of our competitive offering but also because of the significant environmental benefits that our module technology offers.

First Solar's PV technology enables customers to decouple their business growth from emissions, water use and waste generation.

As PV solar continues to become a dominant power-generation technology, we expect that choosing low carbon solar will become increasingly important for our customers, similar to what is already happening in France.

I'll now turn the call over to Alex who will provide more detail on our third quarter financial results and discuss the updated guidance for 2017.

Thanks, Mark.

Before providing an update on our performance for the quarter, I'd like to highlight our 2017 Analyst Day, which will take place on Tuesday, December 5, beginning at 10:30 a.m.

Eastern Time.

The event will be webcast live and will include an update from our executive management team on our strategic priorities related to our technology and capacity road maps, progress in making key markets and our latest business outlook.

There'll be a link to a live webcast to the event available at the Investor Relations section of our website, and our executive management team looks forward to the opportunity to provide you with the latest updates on our business.

Turning on Slide 10.

I'll start by discussing our third quarter operational highlights.

Module production increased slightly in the third quarter to 527 megawatts DC, an increase of 3% from the prior quarter.

Production was 32% lower year-over-year, resulting from Series 4 lines ramped down to make way for Series 6 production.

Capacity utilization, which excludes the lines taken out of service, was 98% compared to 99% in the prior quarter and 97% in Q3 of the prior year.

The full fleet conversion efficiency improved to 17% from the third quarter, a 10 basis point increase quarter-over-quarter and a 50 basis point improvement year-over-year.

Module conversion efficiency on our best line was unchanged versus the prior quarter at 17% but improved 40 basis points year-over-year.

As we've indicated previously, we expect the Series 4 fleet average efficiency to level off near our current 17% efficiency as future technology improvement and investment are focused on Series 6.

Continuing to Slide 11.

I'll discuss some of the income statement highlights for the third quarter, including some non-GAAP measures such as adjusted operating expenses, adjusted operating income, non-GAAP earnings per share and free cash flow.

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

Net sales in the third quarter were $1.1 billion, an increase of $464 million compared to the prior quarter.

The increase in net sales results primarily from the sale of the California Flats and Cuyama projects as well as higher third-party module volume.

Selling the California Flats project was an important milestone to achieve our financial guidance for the year, and in Q3, we recognized nearly 70% of the revenue and margin on the project.

We'll have minimal revenue recognized in the fourth quarter and project activity will pick up again in 2018 as we construct the remainder of the project with the Series 6 module.

Cal Flats is another example of the growing corporate demand for solar power, with 130 megawatt AC of the project offtake contracted to Apple, this represents one of the largest solar projects supplying power to commercial end user.

As Mark mentioned, we're in discussions with corporate customers on a number of additional utility scale opportunities and provide more details at our upcoming Analyst Day.

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, increased to 72% from 63% in Q2.

Third-party module sales increased to 300 million in Q3, up from 228 million in the prior quarter.

Gross margin improved to 27% in the third quarter from 18% in Q2.

This increase results primarily from the higher gross margin on the California Flats and Cuyama projects in Q3.

The gross margin of our component segment improved slightly to 18% in Q3 versus 17% in the prior quarter.

Adjusted operating expenses, which exclude restructuring asset impairment charges, were $84 million in the third quarter compared to $79 million in Q2.

The sequential increase in operating expense is primarily due to higher plant startup costs associated with ramping up Series 6. We expect plant startup costs to increase in the fourth quarter as Series 6 work in Ohio, Malaysia and Vietnam intensifies.

Year-over-year, our OpEx, excluding restructuring of plant startup, decreased by 23%, highlighting the impact of restructuring assets undertaken last year.

Restructuring and asset impairment charges to accelerate our Series 6 transition were less than $1 million in Q3 compared to $18 million in the second quarter.

Excluding restructuring-related items, adjusted operating income in the third quarter was $208 million compared to $32 million in Q2.

The increase in adjusted operating income was primarily due to higher revenue and improved gross margin.

On a GAAP basis, our operating income for the quarter was $207 million.

Tax expense is $8 million in Q3 and includes an $11 million benefit from the expiration of the statute of limitations on various uncertain tax positions.

This compared to a $40 million tax benefit in Q2, which includes a $42 million discrete tax benefit we've gotten from the acceptance of our election to change the tax status of a foreign subsidiary.

Third quarter EPS was $1.95 on a GAAP and non-GAAP basis.

And this compares to Q2 GAAP EPS of $0.50 and non-GAAP EPS of $0.64.

Next turn to Slide 12 to discuss select balance sheet items and summary cash flow information.

Our cash and marketable securities balance ended the third quarter at $2.7 billion, an increase of approximately $490 million from the prior quarter.

Our net cash position increased $467 million to $2.4 billion.

The increase in our cash balance is primarily related to cash received from the sale of the California Flats and Cuyama projects.

Net working capital in Q3, which includes the change in noncurrent project assets and excludes cash and marketable securities, decreased by $415 million.

The change is primarily due to a decrease in project assets from the sale of the California Flats and Cuyama projects.

Total debt at the end of the third quarter was $344 million, an increase of $23 million from the prior quarter.

The increase resulted primarily from issuing project-level debt for a project in Australia.

And as a reminder, essentially all of our outstanding debt is project-related and will come off the balance sheet when the project is sold.

Cash flows from operations were $581 million in Q3 versus cash flows used in operations of $168 million in the second quarter.

Third quarter free cash flow is $484 million compared to negative free cash flow of $272 million last quarter.

Capital expenditures were $98 million compared to $105 million in the prior quarter.

The improved operating cash flow and free cash flow in Q3 was driven by the aforementioned project sales.

Turning to Slide 13.

I'll review our updated full year 2017 guidance.

With the sale of the California Flats and Cuyama projects in Q3, the largest components of our guidance for the year are now complete.

Looking ahead to the fourth quarter, we anticipate the majority of our revenue to be generated from third-party module sales.

We expect minimal system sales in the U.S. as the remaining revenue recognition on the second phase of California Flats will largely occur in 2018, and as we look to optimize other projects in our systems portfolio after Series 6. In terms of international projects, the sale of the small Japanese project originally included in our 2017 guidance is now expected to close next year.

Regarding our project portfolio in India, our guidance continues to assume that we enclose a portion of these projects sales in the fourth quarter with the remainder in 2018.

With that context, I'll discuss the updated guidance ranges in some more detail.

Firstly, as it relates to net sales, we're leaving our forecast unchanged.

We've updated our gross margin guidance to approximately 18%.

Our non-GAAP operating expense range is unchanged, but we've lowered the high end of our GAAP operating expense range by $10 million as we expect lower restructuring expenses.

Operating income guidance has been raised as a result of both the improved growth margin forecast and from the lower restructuring expenses.

The flow-through impact to EPS is to raise our GAAP EPS midpoint to approximately $2.20 and our non-GAAP EPS to $2.50.

Our non-GAAP EPS assumes a full year tax benefit of between $25 million and $30 million, which is unchanged from our prior assumption.

In comparison to our initial expectations coming into 2017, our year-to-date performance reflects not only the strength of our systems and module businesses but also tremendous execution during the course of the year.

Relative to our non-GAAP EPS of $2.86 for the 9 months ending September 30, the $2.50 midpoint of our full year guidance implies a loss in the fourth quarter of around $0.35.

There are several factors impacting the fourth quarter which accounts for this loss.

Firstly, plant startup levels in Q4 will be much higher than in Q3.

And while this expense is necessary to ramp up Series 6, it does present a transitional impact.

Secondly, as mentioned previously, we expect minimal systems business in Q4 due to project timing and as we optimize projects for Series 6. We continue to have a strong project pipeline and anticipate selling a number of projects in both the U.S. and international markets in 2018.

And lastly, with Series 4 production taken offline earlier this year to support the Series 6 transition, lower volume has also impacted sales.

And all of these factors combined to impact our fourth quarter, but our full year earnings remain strong.

Finally, we've left our net cash, operating cash flow, CapEx and shipment forecasts unchanged.

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

We had solid financial results for the quarter, driven by the sale of our California Flats and Cuyama projects.

Our net sales were $1.1 billion, and EPS was $1.95.

Our ending net cash balance is $2.4 billion, an increase of $467 million from the prior quarter.

We raised the midpoint of our non-GAAP EPS by $0.25 to $2.50.

Our Series 6 transition is progressing well, with the installation of the front end of the line in Perrysburg nearly complete and with the arrival of the first coater in Malaysia.

Our net bookings are strong and exceeded 4.5 gigawatts, and our mid- to late-stage opportunities were resilient and currently stand at 6 gigawatts.

And with that, we conclude our prepared remarks and open the floor to questions.

Operator?

(Operator Instructions) And we'll take our first question from Philip Shen with Roth Capital Partners.

What are the key factors influencing, whether or not you book the remaining gigawatt of Series 4?

Can you describe also the terms of the Series 4 contracts for delivery in '18?

Our checks are coming back with -- that you guys have meaningful cancellation fees.

And then finally, given the flexibility of some of your contracts to ship either Series 4 or 6, what is the potential number of megawatts of Series 6 modules that could be sold in '18?

All right.

So I'll try to take those as best as I can.

First off, as it relates to -- decision points as it relates to continuing the production of Series 4, we're continuing to evaluate the options and alternatives that we do have.

One of the primary reasons which we said is we won't do anything that's going to compromise our transition to Series 6. First and foremost, that's the most important thing that we need to do.

And any decisions we make, we'll make sure that we preserve the original committed to Series 6 production schedule as we highlighted for 2018 and 2019.

As we've created optionality within these contracts, it's given us the flexibility, as we indicated.

About the half of the volume that we have booked allows us to either fulfill those contracts with Series 6 and Series 4. If you take the 4.5 gigawatts in net bookings, it would highlight then that there's a meaningful portion of that, that would be -- potentially could be served with Series 6, which would then give you an indication that most of the volume for -- through 2019 could be contracted with Series 6, assuming we made -- did not make a decision to extend Series 4 production.

Some of the contracts also go out to 2020.

So I'm going to make sure that, that's highlighted as well.

The volumes go up that far.

Also, though, remember that 2018 volume for Series 6 largely was consumed within our own development assets already.

So most of that Series 6 shipment profile is mainly going to sit within 2019 and into 2020.

Have we put provisions within those contracts especially those that have mainly focused on Series 4 but also -- and we've incorporated similar provisions in our Series 6 agreement.

As we book that far out, we do want to make sure that there's -- the contracts are an obligation that both parties perform.

There is implications for us if we don't perform, and there's implications to our customers for whatever reason if the contracts or the purchase orders were terminated.

So it's protecting both people's interest to ensure long-term fulfillment against those contracts.

Again, you were going out much further on module sales than we would historically, and so we have included certain provisions with our contract that would protect both our interest in it, and I guess we'll also make sure that our customers' interests are protected relative to our ability to perform and deliver against those obligations.

Great.

That's helpful.

And if it's possible, is it -- can you give us a sense for the gigawatts in '18, '19 and '20?

Is it possible to give us a sense of that mix at all, Mark?

No.

We're not breaking out that detail.

I think the way to look at there's 7.4 gigawatts now that as you think about the balance of this year going out into '18, '19 and '20, so that's a tremendous amount of volume that has been contracted at this point in time.

As we said, the 3.6 that has been originally identified in our production plan, that is sold through at this point in time.

So that entire volume has been sold through.

And if you remember, this year we'll end up producing another -- the run rate we've indicated about 500 megawatts this year, and then what we had indicated for next year was 1 gigawatt of Series 4, effective in 1 gigawatt of Series 6. So that incremental gigawatt that we could produce of Series 4 next year could take Series 4 production up to 2 gigawatts next year.

Look, we'll have to make a decision, as we indicated in our prepared comments, by the end of the year.

We're looking at various events that could inform our views around that and the ability to fulfill those contracts and what's the best product mix to do that with.

And I would expect to have a lot more information as we indicated in our prepared remarks in the Analyst Day in the first part of December.

And we'll take our next question from Jeff Osborne with Cowen and Company.

Yes, if I heard you -- Alex, if I heard you right, I think on the startup cost, you mentioned that there would be a Vietnam-related startup cost.

Can you just talk about what that's for?

I think we talked about at our last call, we have an old factory in Vietnam that was built but never commissioned.

And as part of the optionality that we're keeping around continuing our Series 4, we are moving to use that factory for Series 6. So as we do that, there'll be startup costs associated with bringing that up in the same way they would be with taking the existing Perrysburg and Malaysian facilities over from Series 4 to Series 6.

Yes, I think as Alex as indicated, what we said in our last earnings call, that our effectively third Series 6 plant was going to be in Malaysia.

We've made a decision now to move that to Vietnam.

And what that provides now is this 1 gigawatt of optionality that we're referring to around on Series 4, we could continue to produce more Series 4 in Malaysia for an extended period of time.

That's the decision that we're still evaluating.

But making that decision to go to Vietnam has freed up some optionality around Series 4 production in KLM or Malaysia.

And we'll take our next question from Brian Lee with Goldman Sachs.

I had 2. So first off, given the record bookings, is there a strategy that involves accelerating CapEx in Series 6 as opposed to just keeping Series 4 online longer or maybe potentially doing both in a parallel track?

Just thinking -- or just trying to get a sense of how you're thinking about the Series 6 out year ramp given the initial demand seems to be pushing lead times out here.

And then, just as a follow-up, I'll squeeze this is now, on the margin trajectory, if we look at components, it's been pretty consistent here in the past couple of quarters in the high teens, which makes sense given the pricing stability in the market.

If we take your prior comments around end-of-life margins for Series 4, it sounded like the first half of 2018 we'd see much lower gross margins than where you're tracking at today.

If you do the 1 gigawatt optional Series 4 volume moving through the next couple of years, is it fair to assume, though, that 1 gigawatt optional volume would be in the range margin-wise of what you're seeing today as opposed to the end-of-life commentary you've made in the past?

I'll take the Series 6 acceleration comment and let Alex take the margin comment.

The Series, 6 at least in the near term, as we roll through what we previously highlighted almost a year ago I guess when we did our guidance call, that we roll out and ramped to about 3.5 gigawatts or so of production by the end of '19 with an exit rate that would be that number, call it 4 gigawatts.

The ability to do anything within that time frame is largely constrained by a lead time from our equipment manufacturers.

So anything to really accelerate Series 6 within that window, call it, over the next 8 to 10 quarters I guess, it would be very, very difficult.

But when you go beyond that, as we think about into 2020 and into 2021, there could be some optionality and some decisions that we could make around acceleration, but the problem is we -- that's why there's kind of simultaneous equations here, but the constraints still to get to Series 6 is to ramp down to Series 4. And so we have to play through that equation and that mix.

We've got to solve on what we think is the right production profile for Series 4 and then that will sort of give us the parameters and the goalpost, I guess, that we can make a decision around Series 6. So there's multiple constraints.

Near term, it's the equipment lead time.

Alternatively, there is also the transition from 4 and how much longer we choose to continue to run Series 4 in order to potentially look at options around timing of accelerating Series 6, but to the extent there was optionality would be out in the 2020 time frame.

Yes, and Brian, on the margin point, so we have previously indicated that we are not spending money on improving the efficiency of Series 4. Our competitors are not standing still, and we expected the Series 4 to be at its most challenged just before we start production.

So if we move into '18 under the current plan, competitors won't stand still, and at that point, we would have our most challenging gross margins.

As Mark mentioned in his prepared remarks, we have seen an acceleration of procurement in the U.S., and we've also seen a relatively tight Tier 1 supply-demand market globally based on the current dynamics, mostly in China, and that's had a stabilizing impact on pricing in the near term.

So in the near term, that's why you're seeing margins hold up.

I would also say that if we do extend the Series 4 production beyond our current plan in 2018, we would only do so if we have acceptable margins.

So I think you can assume you would see a similar margin profile going forward into 2018.

And we'll take our next question from Colin Rusch with Oppenheimer.

Can you talk a little bit about the pricing mechanisms in these contracts?

I appreciate that you've got [keys in it] for both sides, but is pricing fixed on these forward bookings here?

And how should we think about that?

Yes.

So they're very similar to bookings that we've had historically, right?

So there is a fixed price associated, a fixed price from Series 4. There's a fixed price from Series 6. Now we do have -- this will primarily relate more to Series 6 because there's not going to be a significant difference of bin availability for Series 4, but we do have provisions in our contracts that would give us what we refer to as [bin adders] or price adjusters, such that, if we ship them a higher-watt panel, then there's a price increase associated with that.

But these are fixed prices that go out for an extended period of time, as we've indicated.

They would potentially have some bin adders . And more of that will be in the Series 6. There will be some in Series 4 as well.

But think of them as firm fixed prices, obligations by both parties to perform and implications to the extent that those parties do not perform.

And we'll take our next question from Krish Sankar with Bank of America Merrill Lynch.

This is Chirag Odhav on for Krish.

Can you guys generally see higher module prices leading towards slower IRRs for the project business?

Do you expect this to make things a little less attractive for our future products -- or projects?

So if I look at the IRRs we're seeing in the market today, I think you have to look at the total underwriting assumptions in it.

So the IRR headline number itself has stayed I think relatively resilient despite potential headwinds around tax reform and interest rates and other things.

So I think we're seeing IRRs stay relatively stable.

If I look to whether the module pricing is impacting those, clearly, the model is pretty simple, right?

You've got your input costs, your lower revenue stream and your return on capital.

So if you change one of those, you're going to have an impact on one or other of the remaining 2. So I think clearly, there's going to be some impact.

I think if you look at the increase in pricing we are seeing on the module side, it is not significant to the point where it's going to have hugely detrimental impact on IRRs.

And depending on when your project was bid and the assumptions that you used at the time around forecasts, interest rates, yield curves, cost of tax equity under our other operating expenses, motion curves and other things, I think the general level of increase in module pricing is still allowing viable projects today.

Yes.

I think the thing to remember, too, is that I think the module represents something close to about 20% of the overall LCOE.

And so it is a small component relative to the other 80% when you aggregate everything else up.

So it shouldn't have a significant impact, but the reality is what was the assumption that was used when those PPAs were bid.

If there were assumptions that module prices were going to fall very quickly, and that there was an assumption of a continued oversupply industry that, therefore, people would be selling at or below cash cost in order to liquidate inventory, where pricing is now settling to could have a more impactful impact to somebody's return expectations because of a very aggressive assumption they had when they bid into a particular power price.

But on a normalized basis, if you're just looking at the delta from where module prices were at the low to where they are now, that's not going to have a significant impact.

But if you were taking a low point and then further assuming price erosion against it, you could have a more significant impact from that standpoint.

And we'll take our next question from Vishal Shah with Deutsche Bank.

Mark, what percentage of your bookings historically had been Module Plus bookings?

I appreciate the comment you made about negotiations you've had with some of your existing customers.

So should we assume 30%, 40% of these bookings would eventually get converted to Module Plus?

Or could that be a smaller number?

And then also what percentage of your bookings are with some of the utility customers and corporate buyers versus some of the smaller developers and distributors that could potentially negotiate, renegotiate on pricing?

So in terms of -- I guess what I'll throw in into the bucket for when you say Module Plus, I'll throw in kind of EPC and O&M services into that discussion.

As we indicated in the call, there are -- there's multiple hundreds of megawatts of opportunity against the bookings that we recognized that we could capture additional services for O&M and EPC in particular.

And what's happened is that there's, obviously, a desire.

We're constrained.

And there is a desire to lock up on module prices as quickly as possible.

We're still in negotiations with some of our customers that would include additional scope beyond just the module, but it's a relatively significant number, but there's a lot a work still to be done to determine whether or not we can capture something beyond just the module side of the equation.

So we'll work through that.

As it relates to the C&I bookings, what we said in the call, I guess, 400 megawatts or so of module sales that is related to a customer -- a commercial industrial customer, there's 800 megawatts of late-stage negotiations that's against the 2 gigawatts of systems business that we highlighted that is for C&I.

That segment of the market is relatively robust.

We're seeing a lot of activity there and really it plays strongly to First Solar's capabilities and brand strength, bankability, balance sheet strength.

The buying decision-maker there I would align more to a utility who's looking for utility-owned generation.

We're not selling into a developer who's only looking to develop and then sell on the asset as quickly as possible and not thinking about total life cycle, cost of ownership and performance and those types of things.

We're buying -- we're selling to the much more sophisticated informed buyer that can clearly understand the differentiation capability that First Solar can provide them.

So it's a very exciting segment of the market, and one that we expect to continue to grow.

Vishal, you mentioned small developers that could renegotiate, I just want to reiterate that the contracts we're signing today are firm obligations on both sides with penalties for failure to perform on both sides.

So regardless of size of the developer, we have taken bookings with contractual terms that we think will hold on.

On both sides, there have been security deposits paid.

So I don't see a significant risk of renegotiation.

And we'll take our next question from Edwin Mok with Needham & Company.

First, mainly -- my question is around bookings, right?

So I guess first, if I remember, you guys are targeting around 2 gigawatt next year and maybe another 1 in [3/01/19], so assuming that you're fully booked out for 2020, since you've mentioned that your bookings are stretched out to 2020, and how -- maybe the other way to ask is how linear is that booking?

And then just quickly in terms of converting those projects from third-party module to (inaudible) system cells, does it mean that you are successfully doing that, that could be pretty meaningful upside to this 1 [cumulative] target, per-year target that you laid out?

Sure.

So I guess on the bookings side, and then I'll let Alex take the second one, the -- well, if you take -- what we said in the -- in November of last year, we'll have about 2 gigawatts in '19, 2 gigawatts in '18, about 3.5 in '19 and then you got a run rate number that gets you to about 4 gigawatts into '20, all right?

So you've got 5.5 plus 4, so that's 9.5 over those years, 2, 3.5 and 4. Plus you still have 500 or so this year, and so you can add that into that number, so you're starting to get to a number north of 9 gigawatts.

And what we said is that we contracted 7.4 gigawatts, is what we still have left, all right?

So that would tell you that assuming nothing else changes, assuming we do nothing on Series 4 in adding incremental capacity, assuming we don't do anything to try to accelerate some Series 6 into 2020, then we've got somewhere in the range of 2 gigawatts left to sell over that horizon.

I mean, that's kind of what the pure sense of the math will tell you, and that's the kind of backdrop of analysis that we're looking at right now to look at various scenarios and determine what is the best thing to do to best meet the needs of our customers as well as drive to the optimal financial results.

And the one thing I want to continue to remind people on is that variable contribution margin flow-throughs flows through to accretive EPS expansion, right?

So when you -- we've said before, we showed you a slide last year.

I refer you to those slides we showed in the last guidance call, right, that we can leverage fixed costs in our manufacturing but, more importantly, in our OpEx.

So as you think about that contribution to volume flow-through, it's going to flow through to a very attractive EPS impact, not only because of the fixed cost structure and their manufacturing OpEx, but as most of you also know, we have a very efficient tax rate.

And so a lot of that incremental contribution margin flows through almost 100% to EPS expansion.

So we're looking at all options.

It's very complicated.

A lot has happened very quickly, and we'll share more of that information and what our views are around that in the Analyst Day in December.

And Edwin, on the 1 gigawatt of systems business and the upside there.

So we've mentioned before that we're tracking to about 1 gigawatt a year on average of systems business, of self-development and third-party EPC work.

It's possible that in 2018, we may be under that average as we are looking to optimize our self-development platform for Series 6. However, in the near term, if you look today, towards that 1 gigawatt, we're about 2/3 of the way there for 2018 today.

There are a lot of opportunities, as Mark said, to bring that 2018 number up to 1 gigawatt given this recent accelerated procurement dynamic where modules are being locked in ahead of other services around EPC and O&M.

If we can bring that beyond 1 gigawatt, we're very happy to do that, and we'll continue to work those opportunities in both '18 and then beyond into '19 and '20.

Yes, and I think the other thing too is that this is, obviously, it's a very impactful decision, strategic, long term in nature.

I think our assessment around that and the rationale on how we think through that to me is better communicated in a form, such as the Analyst Day, right?

So we can get more detailed and thorough discussion, quality of discussion around that.

And so I know everybody is going to continue to sort of ask the same question, but at this point in time, as I -- as we've indicated, we are where we are.

We'll give you more information in December.

We'll take our next question from Cindy Motz with Williams Capital Group.

I was going to rephrase that '18 guidance another way, but I guess we will get it at the Analyst Day.

So but would it be fair to say that just everything you've been talking about, Mark, and Alex, your -- just the demand you're seeing from utilities and, I guess, C&I.

And I was just going to ask you about community solar, how that's going, but that you're feeling like better about '18?

And then also, in general, I was thinking about the sustainability thing you said that people are coming to First Solar because of that, the sustainability metrics and everything.

Is that just overseas?

Or is it in the U.S. as well?

And then I have one follow-up.

Yes.

First, I guess, I'll take the sustainability question.

First, is that it is really -- it's impactful and it's driving decision-making, and where it really hits and resonates extremely well is with large-scale commercial, industrial customers in particular, right?

I mean, their whole intended objective is, is obviously, to focus on clean renewable energy and not only for their own consumption but they're also looking at that from a standpoint of their supply chain.

So some of our customers are actually making the requirements to have flow-throughed and into their supply chain and using that as a criteria that they're making supplier decisions around.

And so when our sustainability efforts come through and we highlight the benefits that we can provide to the environment, it really resonates extremely well with the number of our large-scale commercial industrial customers.

And sometimes it even gets us into the table into discussion.

It's just that concept, in and of itself, is sort of facilitating some conversations with our customers.

The international side though, we highlighted the tender process that we've done in France, and it is more impactful.

Actually, I was over in Europe recently speaking at a conference and that even some of the other European countries are even looking at that as a criterion.

And I don't -- and I think it may evolve deeper into other regions of the world, and it's something that we think needs to be taken in consideration because when you look at the payback period relative to our technology compared to other technologies, it's a much quicker, faster payback.

And if you really focused around reducing greenhouse gas emissions, it's a better solution or way of doing that.

On the -- I'm sorry, on the utilities C&I community solar demand, I think, as we mentioned in the scripted comments, is over 800 megawatts of projects with corporate customers in that list.

So as Mark said, the sustainability is very helpful and we're seeing that drive for C&I interest, significant interest across the board for the utility as well.

There's a lot of interest in getting solar in.

A lot of people have targets for end of decade.

And across the IPP space and developer space we're seeing interest to getting ahead of an ITC expiry.

So across the board, utilities and C&I community solar we're seeing a lot of activity and lot of driving force.

And we'll take our final question from Joseph Osha with JMP Securities.

If you look at your production plans for 2018, 2019, what you just said about Vietnam, the 3.5 gigawatts and so forth, it seems as if there might be room for a greenfield fab in here somewhere.

I'm just curious as to your reaction to that statement and whether there might -- that might be part of the planning process.

Yes, look, we're looking at all options.

And as we go further out into the horizon, that type of option does come into play more so than near term.

Near-term is the fastest way to get more Series 6, just use the existing brownfield that we already have.

But as you go further into the horizon, a greenfield or a modified greenfield, it could be an expansion to one of our existing facilities.

It could be a new building to -- in Vietnam or could be a new building in Malaysia or could be a new building in Perrysburg, right?

So it's an attachment to an existing manufacturing operation.

It would probably be the priorities on how we think through it.

But that, clearly, that's part of the decision-making that we are evaluating.

But again, those would be situated further out in the horizon in 2020 and beyond.

And this concludes today's question-and-answer questions as well as the First Solar's Q3 2017 financial results call.

Thank you for your participation, and you may now disconnect.