第一太陽能 (FSLR) 2018 Q1 法說會逐字稿

Good afternoon, everyone, and welcome to the First Solar's First Quarter 2018 Earnings Call.

This call is being webcast live on the Investors section at First Solar's website at investor.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, Ashley.

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

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

A copy of the press release and associated presentation are available on our website at investor.firstsolar.com.

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

Mark will begin by providing a business and technology update.

Alex will then discuss our financial results for the quarter and provide updated guidance for 2018.

We will then open 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 a few cases where we report non-GAAP measures such as free cash flow or non-GAAP EPS, we have reconciled these 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 is now my pleasure to introduce Mark Widmar, Chief Executive Officer.

Mark?

Thanks, Steve.

Good afternoon, and thank you for joining us today.

Moments ago, we issued a press release announcing that First Solar plans to construct a new Series 6 module manufacturing facility in Ohio, which will significantly increase our domestic production capacity.

This decision is in response to the continued strong demand for our Series 6 product and allows us to better address the growing U.S. solar market.

Furthermore, it is a strong testament to the competitiveness of U.S. manufacturing within the global landscape.

As highlighted on Slide 4, this greenfield factory will have a nameplate capacity of 1.2 gigawatts and will be built essentially next to our current manufacturing operations in Ohio.

Combined with our existing Series 6 factory in Ohio, which has a nameplate capacity of 600 megawatts, this additional facility will bring the total planned Series 6 production in the U.S. to 1.8 gigawatts and further solidify our position as the largest U.S. solar module manufacturer.

We expect production to begin in late 2019 and the factory to be fully ramped by the end of 2020.

The total capital expenditures to construct this facility are expected to be approximately $400 million and will be incurred over the next 2 years.

The site selected could accommodate a second factory of equal size, and we will continue to monitor market conditions and our relative competitive position as we evaluate the potential for adding additional capacity.

There are 3 major benefits to building a new fully integrated manufacturing facility in the U.S.: firstly, recent U.S. tax reform and the favorable local and national business environment enhance the economic rationale for U.S. manufacturing; secondly, the lower labor cost per watt of our Series 6 module reduces the labor arbitrage benefit of manufacturing in Malaysia or Vietnam relative to the U.S.; lastly, by locating the factory near our existing U.S. manufacturing facility, we will be able to efficiently leverage the capabilities and know-how of our experienced R&D and manufacturing teams.

In terms of local economic impact, there will be a benefit from the capital spend to construct the facility, which is in addition to the approximately 500 new high-quality permanent manufacturing jobs that will be created directly by our operations.

We also expect to significantly grow our local supply chain to support the higher production volume and, thereby, create additional new jobs and economic benefits.

Furthermore, this growth will create economies of scale for our supply chain partners to enhance their cost reduction road maps.

Turning to Slide 5. An important item to highlight is that our new factory announcement will not have any impact to the previously announced Series 6 road map as we exit 2020.

As shown on Slide 5, this new U.S. factory Series 6 with this U.S. new series -- U.S. factory Series 6 global nameplate capacity is expected to increase to 6.6 gigawatts.

In combination with approximately 1 gigawatt of Series 4 capacity, we expect to have 7.6 gigawatts of total nameplate capacity as we exit 2020.

Relative to the production plan that we provided at our Analyst Day in December, today's announcement, combined with the recent re-start of 2 Series 4 lines in Ohio, is expected to add an incremental 1.2 gigawatts of supply over the next 3 years.

We will provide additional details on the progress of the latest Series 6 factory on future calls.

Turning to Slide 6. I'll next discuss the progress we have been making to ramp our Series 6 factories.

In terms of manufacturing readiness, which captures elements such as building preparation, tool installation, staffing and other activities, we continue to make excellent progress, and we scored our readiness as green across all 4 of our factories.

The most notable achievement to highlight is the start of production at our Ohio factory at the beginning of this month.

To put this achievement in perspective, 18 months ago, when we announced the decision to accelerate our transition to Series 6, we were confident in the journey we had embarked upon.

However, the new equipment and manufacturing process that needed to be designed and validated we knew the journey would be challenging.

The fact that we have been able to meet our initial target of mid-2018 production and begin shipping modules to projects is a tribute to the capabilities of our associates and First Solar's commitment to delivering results.

Next, in Malaysia, we achieved a significant milestone earlier this week as we produced our first complete module.

The factory continues to move steadily closer to the start of production in Q3 and is aided by the learnings from the initial ramp in Ohio.

We are likewise making good progress with both Series 6 factories in Vietnam.

Front end in line of the first factory is now 90% installed, and the construction of the second factory is continuing on schedule.

While we are pleased with the manufacturing readiness progress, we are working through a number of issues related to our manufacturing ramp.

For this reason, we have scored the 3 associated metrics of wattage, throughput and product readiness as yellow.

The first thing to note is that the issues we are having are normal occurrences for this phase of factory ramp, and we are proactively addressing them.

Related to module wattage, we are encouraged by the steady progress and ongoing improvements that we have been making since we started the line.

The current production entitlement yields 90% of the modules being produced at or above 400 watts with the top end of the distribution at 420 watts.

Over the next 90 days, we see a clear path to an average production bin of 415 watts and a top bin of 425 watts.

Furthermore, the early module wattage gives us confidence in our long-term road map, which takes us beyond 425 watts per module as we discussed at our last Analyst Day.

Throughput, which measures the number of modules produced each day, is the area where we are currently placing the greatest focus and working to improve equipment availability in order to increase production levels.

Notably, the front end of the manufacturing line, which includes the most complex and unique processes, such as deposition and laser scribing, is performing well.

However, the less technically challenged back-end assembly processes are where the current constraints are, and therefore, we are placing the greatest focus on this portion of the line.

As we indicated at the onset of the transition, we believe the greatest potential risk was not whether the core Series 6 module technology would function properly but whether there would be scheduled risks given the 4 factories will be ramping in succession under an aggressive time frame.

To put the schedule risks in context, it is not a matter if we will reach the expected output levels but rather a question of when we will get there.

To the extent that some of our existing challenges persist, Series 6 supply in 2018 could be at the low end of our expected range.

As originally planned, we believe that any near-term shortfalls in Series 6 can be managed and resolved effectively given flexibility that we have in the module delivery schedules to our own self-developed projects.

Overall, we are encouraged by how far we have come on this transition.

But due to the complexity of the undertaking, the potential for near-term production delays still exist through the ramp period.

The third manufacturing ramp metric provided is Series 6 product readiness, which captures product testing requirements, whether from a customer or other entities.

While the product is meeting the design intent on all fronts, there is scored as a yellow based on completion timing.

We anticipate that certification may slip a few weeks past our original target.

However, we remain confident that it will not be a question of whether we obtain these certifications but rather a question of time before it is completed.

One other note on product readiness is that as our EPC customers and structures ecosystem partners continue to familiarize themselves with Series 6 they have continued to road map new innovative mounting approaches and to request module capabilities that enable lower total systems cost.

First Solar is evaluating Series 6 frame design variants that can also bring additional mounting application capabilities in response to this customer's input.

In summary, while we're working through typical ramp related issues, we are very pleased with our progress to date.

The execution to reach this point has been tremendous, and we are encouraged by the manufacturing readiness of our factories and the current module wattage results.

Next, I'll turn to Slide 7 to discuss our bookings activity since our last call.

As I've mentioned previously, our decision to expand capacity was based on the strength of the demand for our Series 6 product.

In the roughly 2 months since our last earnings report, we have contracted 2 gigawatts of additional volume, which brings our total year-to-date net bookings to 3.3 gigawatts.

After accounting for shipments through the end of Q1, we now have a balance of 10.6 gigawatts of future expected shipments at this time.

While this level of visibility -- with this level of visibility into shipments between now and 2020, combined with our mid- to late-stage pipeline, which I will discuss shortly, we feel confident in making the decision to add additional capacity.

The largest single booking among the recent deals we have signed is a 750 megawatt module supply agreement with a leading U.S. developer.

Combined with this separate agreement signed last year, we have now contracted with this customer for over 1.2 gigawatts of module deliveries in 2019 and 2020.

Our other bookings for the quarter were primarily module sales and include deals with various U.S. customers as well as international bookings in Turkey, France and other locations.

With more than 350 megawatts booked in Europe since the beginning of the year and over 580 megawatts booked during the past 12 months, we are making good progress in this region.

Our continued progress in these markets is aided by our low carbon footprint, energy yield advantage and focus on customer engagement.

While not included as new bookings, since our last earnings call, we have signed EPC agreements with 2 customers for a total of 380 megawatts DC.

In each case, a module supply agreement was already in place, and we've now added EPC scope.

The first of these agreements was with Vectren to construct a 60 megawatt DC project in Indiana that's expected to power more than 11,000 households.

Construction of the project is expected to begin in the second half of 2019 with project completion in 2020.

This utility-owned project will utilize Series 6 modules and highlights First Solar's plant design approach, which is tailored to utility ownership values.

We've also signed an EPC agreement with Longroad Energy for a 315 megawatt DC project in Texas that is expected to be completed in late 2019.

We look forward to future collaboration with Longroad as they continue to build out their U.S. development pipeline.

In combination with our previously signed EPC agreement with Tampa Electric, we now have over 620 megawatts DC of contracted EPC projects that we will construct over the next 3 years.

In addition, with over 500 megawatts of EPC opportunities that are not yet booked, we are still in discussions with customers.

Turning to Slide 8. I'll next discuss our mid- to late-stage booking opportunities, which have grown to 8.3 gigawatts DC from 6.8 gigawatts in the prior quarter.

With 2 gigawatts booked during that time period, the actual gross increase in opportunities is approximately 3.5 gigawatts.

In terms of the geographical mix, the largest increases were in the U.S. and Asia Pacific.

The increase in the U.S. opportunities were a combination of both systems and module opportunities with a combination of utilities, developers and corporate customers.

In Australia, several project development opportunities have progressed to the point where they are now included in our pipeline.

The remaining increase in opportunities is made up of potential module sales, primarily in Japan and Australia.

Similar to last quarter, the total potential opportunities include deals that are signed but not yet counted as bookings.

Approximately 1.2 gigawatts of such projects are included in the 8.3 gigawatt total.

And the final booking decision is dependent on the closing of financing our other CPs.

The shipment timing of these projects is also noteworthy as more than 6 gigawatts or approximately 3/4 of the total mid- to late-stage opportunities have deliveries in 2020 or later.

The profile of these potential bookings matches up closely to the timing of the additional capacity that we have announced today and provides good visibility into the demand over the next several years.

Lastly, we saw positive growth in the mix of system opportunities in our potential bookings versus the prior quarter.

Systems project opportunities now comprise 1.9 gigawatts as compared to less than 1 gigawatt approximately 2 months ago.

This increase is a result of the progress we are making in our project development pipelines in the U.S. and Australia.

Taking into account the fact that we have booked approximately 600 megawatts of development business and signed EPC agreements of 620 megawatts primarily since at the beginning of the year, our recent momentum is building and our systems pipeline is encouraging.

With additional mid- to late-stage development opportunities of nearly 2 gigawatts, we have a strong foundation to meet our target of 1 gigawatt per year of systems business.

I'll now turn the call over to Alex who will provide more detail on our first quarter financial results and discuss updated guidance for 2018.

Thanks, Mark.

Turning to Slide 10.

I'll start by covering the income statement highlights for the first quarter.

Net sales in Q1 were $567 million, an increase of $228 million compared to the previous quarter.

The increase is due to the sale of 155 megawatts of projects in India, 15 megawatts of projects in Japan and initial revenue recognition from the sale of the Rosamond project in the U.S.

It's worth noting that although our first Series 6 modules produced will ship to the second phase of our California Flats project, which was sold in 2017, Rosamond is the first project we've sold that will use exclusively Series 6 modules.

As mentioned on last quarter's call, there was a potential for loss in Q1 if some or all of these projects did not close.

But the execution of these deals is a significant contributor to revenue and earnings for the quarter.

First quarter revenue also included $55 million related to the settlement of the California State sales and use tax examination associated with the EPC contracts.

As a percentage of total quarterly net sales, our systems revenue in Q1 was 72% as compared to 39% in Q4.

Gross margin improved to 30% in the first quarter from 18% in the prior quarter.

The higher gross margin was primarily a result of the mix of higher gross profit projects recognized and the settlement of the sales and use tax examination.

On a segment basis, our first quarter systems growth margin was 40%, and our module gross margin was 6%.

The systems segment margin was strong as a result of the India, Japan and U.S. projects sold in Q1.

The decline in the module segment gross margin was due to lower Series 4 ASPs associated with supply agreements that were contracted several quarters ago.

The average ASP of module sales in Q1 is not reflective of pricing on more recent bookings, and we expect module segment ASPs to improve in the coming quarters.

Even as ASPs increase, the module segment gross margin through the remainder of the year is expected to be low due to ramp-related costs that begin in Q2 and because initial Series 6 module production is targeted towards the systems business.

As a reminder, our segment reporting was revised last quarter, and the module segment now includes only modules sales to third parties.

The systems segment includes all revenue from the sale of solar power systems, including the module.

Q1 operating expenses were $99 million, an increase of $2 million compared to Q4.

Plant startup increased by $17 million as a result of higher Series 6 preproduction activities across our Ohio, Malaysia and Vietnam factories.

The higher startup expense is mostly offset by lower SG&A and R&D expense, which was also from a decrease in variable compensation as well as efficient expense management.

Operating profit for the first quarter was $74 million compared to an operating loss of $35 million in the fourth quarter.

The increase in operating profit was a result of higher net sales and the higher mix of systems projects sold.

Other income of $18 million in Q1 as a result of the $20 million gain on the sale of certain restricted investments associated with the reinvestment of overfunded amounts from our module collection and recycling trust.

Income tax expense for the first quarter was $12 million compared to $399 million expense in the previous quarter.

Q4 tax expense is impacted by U.S. tax reform and, as indicated previously, is subject to additional analysis and further interpretation and guidance from government regulators.

We did not record any adjustments in Q1 related to our original provision, and we continue -- expect to continue revising our provisional estimates until we file our 2017 federal tax return later this year.

Earnings per share for Q1 was $0.78 compared to a loss per share in the fourth quarter of $4.14.

Adjusted for the impact to the U.S. tax reform and the effects of restructuring and asset impairment charges, the non-GAAP loss per share in Q4 was $0.25.

And please refer to the Appendix of the earnings presentation for the accompanying GAAP to non-GAAP reconciliation.

Moving on to Slide 11.

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

Our cash and marketable securities balance ended the quarter at $2.9 billion, a decrease of $110 million from the prior quarter.

Our net cash position decreased by $155 million to $2.4 billion.

The lower cash balance is primarily due to increased capital expenditures in Q1 to support our Series 6 ramp, partially offset by the $102 million received from the reimbursement of overfunded amounts from our module collection and recycling trust.

Q1 net working capital, which includes the change in noncurrent project assets and excludes cash and marketable securities, increased by $150 million.

The change was primarily due to an increase in inventories, accounts receivables and lower accrued expenses from variable compensation payments.

Total debt at the end of the first quarter was $438 million, a net increase of $44 million from the prior quarter.

The increase was also from issuing project-level debt in Japan and India.

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

Cash flows used in operations in Q1 were $45 million due to variable compensation payments and increases in inventory, partially offset by project sales.

In Q1, customers of our international projects received approximately $60 million of liabilities related to these transactions.

And if these sales have been structured differently such that the project debt was not assumed by the buyers, operating cash flow would have been $60 million higher.

The following simplified example helps to illustrate this point.

If we were to sell a fully constructed unlevered asset for $100, we would see a $100 increase in revenue, $100 increase in net cash and $100 increase in operating cash flow.

If we sell that same asset for $100 but leverage with $80 of nonrecourse project level debt, we likewise see $100 increase in revenue and $100 increase in net cash.

However, the $100 increase in net cash is comprised of a $20 increase in cash and an $80 reduction in debt, which is assumed by the customer.

And significantly, this means there's only a $20 increase in operating cash flow in this example.

Therefore, the same asset sold to the same economic profile and the same revenue and net cash impact can show a significantly different operating cash flow profile, depending on the transaction structure.

Historically, we've sold either constructed unlevered assets or assets where the sale have occurred prior to construction and, therefore, prior to the drawing down a significant debt on the project.

Despite the visual effects we're showing, reduced operating cash flow as we develop assets and hold them on balance sheet for some portion of the construction, especially in international markets, there are benefits to utilizing nonrecourse project leverage, including managing FX risk, optimizing construction equity working capital and ultimately, optimizing overall project value upon sale.

Finally, capital expenditures were $178 million in the first quarter compared to $199 million in the prior quarter.

Our spending on Series 6 decreased slightly.

Depreciation and amortization expense was $24 million in Q1.

Continuing on Slide 12, I'll discuss our 2018 guidance.

Our Q1 earnings provided a positive start to the year, but we're leaving our income statement guidance unchanged for the time being because of risks associated with project sales timing and potential cost impacts associated with steel and aluminum tariffs.

The primary update we're making to our guidance of project timing pertains to the sale of our Beryl project in Australia, which was originally anticipated in 2018 but we now expect to complete and recognize revenue on in 2019.

While the revenue and associated earnings for this project are moving to 2019, we're not lowering our income statement guidance due to offsetting items in our Q1 results.

Separately, our Ishikawa project in Japan, which is included in our guidance for the year, experienced weather-related construction delays in recent months, which increases the risk of completing a sale of the project this year.

We've identified an action plan to mitigate the impact of these delays, however, given the size of this projects and the expectations that the sale and, therefore, initial revenue recognition will occur near to or at COD of the project, we believe it's prudent to highlight the risk.

If the project sale does move into 2019, there would not be any change in the expected project economics, however, we could fall to the low end of our 2018 revenue earnings and cash guidance ranges as a result.

Another potential risk to our 2018 guidance is higher steel and aluminum costs due to recent U.S. tariffs.

The tariffs have a direct impact on certain items such as posts and torque tubes used in construction of systems projects as well as on the frame on our Series 6 module.

We're also seeing indirect impact to these tariffs through high commodity prices.

We continue to evaluate potential impact to these tariffs and are also actively pursuing mitigation strategies in order to minimize the impact to the year.

As it relates to distribution of earnings over the remaining quarters of the year, we expect Q2 to be approximately breakeven.

But similarly to our guidance for Q1, this could change either positively or negatively, depending on the timing of the closing of certain project sales and the closing of the 8point3 transaction.

But part of the reason we expect Q2 earnings be lower is the Q2 total OpEx is expected to be highest of any quarter of the year.

This is due to both startup expenses peaking next quarter as well as the timing of some expenses pushing from Q1.

We expect Q4 to be the strongest quarter of the year due to higher systems sales.

We're lowering our net cash guidance by $100 million to a revised range of $2 billion to $2.2 billion primarily due to the incremental $150 million of capital expenditures to support the new Series 6 factory included in this year's guidance.

Note that the remaining $250 million of CapEx for new factory is expected to be incurred in 2019.

Operating cash flow is $100 million lower due to the updates we're making to the Beryl project, which is originally assumed to be sold as an unlevered asset but will now be financed with project-level debt.

This adjustment did not impact net cash as it is an offsetting financing cash inflow.

Lastly, as discussed previously, the structure of project sales transactions can have a significant impact on operating cash flows.

In addition to the structure project sales, the timing of a sale can also have a meaningful impact on operating cash flow guidance to '18.

For example, if we sell a project later in 2018 than anticipated and the project continues to draw down debt financing in intervening periods, the operating cash flow proceeds will be lower, assuming the debt is assumed by the buyer of the project.

Even though the economic substance for the transaction is unchanged, the classification of cash flows may be different, and this can impact to our expectations for operating cash flow for the year.

Finally, turning to Slide 13, I'll summarize the key messages from our call today.

Firstly, as Mark discussed, we're encouraged by the continued strong market demand for our Series 6 technology.

With future contracted module shipments of 10.6 gigawatts and potential bookings opportunities of 8.3 gigawatts, we have strong demand visibility, which enable the decision to expand capacity.

We're excited to add some manufacturing presence in Northern Ohio, and we'll continue to evaluate further expansion in the future, subject to market conditions.

Secondly, we're making progress on our Series 6 ramp, and we reached important milestones this month as we started production and again, the first commercial shipments.

Thirdly, we're pleased with the results for the quarter and the strong project sale execution and effective core OpEx management, which contributed to our earnings of $0.78 per share.

Our net cash position remains strong at $2.4 billion even as we continue to invest at Series 6 capacity.

And lastly, we're maintaining our 2018 earnings guidance and updating our net cash guidance as a result of new capacity expansion announcements and project sale timing.

And with that, we conclude our prepared remarks and open the call for questions.

Operator?

(Operator Instructions) And we'll take our first question from Sophie Karp.

We'll take our next question from Philip Shen with Roth Capital Partners.

It looks like you're making great progress on the Series 6 ramp-up.

When do you expect your first third-party shipments?

If you said it in your prepared remarks, sorry if I missed it.

I know it's always been kind of perhaps back half of this year, but now that you're -- you have a clear line of sight to ship to your own projects, you mentioned something about certifications may slip a few weeks.

Does the first shipment to third parties depend on that certification?

And what was that original deadline?

And where -- what is that date now?

And then shifting to the progress you're making on Series 6, 90% of the modules at more than 400 watts, that's fantastic.

Can you give us an update on what you're seeing in terms of the cost structure progress you're making?

Back at the Analyst Day, I think at year-end '16, you talked about 40% lower than Series 4. So is that very much on target as well?

And then finally, how do you expect the bookings to trend with the success that you're having with the Series 6 ramp-up?

All right.

So a lot there.

I'll try to hit on it.

The -- as it relates to third-party shipments for Series 6, and again, as we originally planned, the vast majority of the early production was going to go into our own self-developed assets, and that's what is happening.

The shipments to the third parties really don't start to happen until, really, the end of the year into the November, December time frame where we really start to see any meaningful volume starting to go into third-party shipments.

So again, we've allowed this optionality and flexibility with our own self-developed assets to absorb the volatility potentially around the ramp and potential indications that, that may have or even the timing of various certifications.

As it relates to the certifications, again, the third-party has actually no impact at all, if anything, going to third-party.

But the other thing I just want to put into perspective is that we have historically, and we will do it again here, effectively do in-field certification.

So we've done this before with Series 4, and we have continued to ship and then do the infield certification.

And so the initial shipments that are going out right now to one of our self-developed assets, we will use that same type of methodology.

So I don't want you to feel like the timing around certification and the fact that it's slipping a little bit really will have any impact -- not only will it have any impact to third-party customer shipments because they're further down the horizon but even to our own self-developed assets, it won't be an issue from that standpoint.

Series 6 volumes are -- the costs side of the equation, (inaudible) for what we're seeing right now with some potential stress points on aluminum, mainly with what's going with tariffs and, in general, commodity price increases, there's a little bit of friction from that standpoint.

So if I look at it, where are we stressed a little bit on our cost per watt assumptions around Series 6 is primarily related to the frame.

Beyond that, I feel very good where we are.

Again, we have to get to high-volume manufacturing, much higher-volume manufacturing where we're in right now, which will continue to help drive costs down as well.

We feel good about that relative to the 40%.

And again, that was an exit end of '16 type of number, and we feel very confident where we'll be with Series 6 once it is fully ramped.

From a bookings standpoint and what that profile should look like, and when you look at it right now, as we indicated in the prepared remarks, the incremental capacity that we now added with the Series 6 plus the decision we announced last quarter to run our 2 lines in Perrysburg for -- actually, we started them because we initially had shut them down, the combination of the 2 of those will give us about 1.2 gigawatts production over '18, '19 and '20, which, if you look at that relative to what we said in the Analyst Day now says that we're -- and our cumulative production of that horizon is about 15 gigawatts.

You pull out 500-or-so megawatts that we ship right now through the first quarter, which says we've got another 14.5 gigawatts of production that will happen over the next several years.

And against that, we have 10.6 gigawatts booked, which means we've got 4 gigawatts to continue to book.

Against that number, I've got 1.2 that is signed right now in our mid- to late-stage, subject to [CPs].

But you started looking a little bit north of 3 gigawatts to have everything sold through in that horizon.

When I look at the north of 8 gigawatts back out 1.2, that leaves me 7 gigawatts of mid- to late-stage to realize 3 gigawatts or so of additional bookings to make sure we sell through all the way through 2020.

A lot of work still need to be done.

But could we be in a position by the end of this year that largely has gotten that accomplished and then starting to put more points on the board beyond 2020?

I mean, that would be the goal, and there's a path to say that, that could happen.

And we'll take our next question from Sophie Karp with Guggenheim Securities.

Just to follow up from the discussion on the bookings, if you continue to book these volumes, how much of embedded pricing risk is there?

Maybe are there other different contract types that you're signing?

And just how much of, I guess, price declines can -- do you anticipate when you think about it?

Well, look, Sophie, the way our contracts are structured, and they have been through the horizon that we've been talking about, building the contracted pipeline, is there are firm fixed prices that also come associated with some form of payment security or downpayment, whether in the form of an LC or cash or something along those lines.

And they're enforceable, and there's obligation for the customer to take, and there's obligations of First Solar to perform relative to those contracts.

There are not any price adders or deltas per se but for price adder for the bin.

So if we contract at a 430 watt module and we deliver a 435 watt module, then there'll be a price adder.

If we deliver a 425 watt module, there will be price deduction.

That's really the only variability in the ASPs.

Other than that, other than the bin that ultimately is -- or delivered to the customer relative to what was contracted, and there's provisions in the contract to allow for deltas on the contracted bin.

It goes up and down.

Other than that, there's really no variability into the contracted ASP.

And we'll take our next question from Ben Kallo with Robert W. Baird.

Can you talk a little bit about guidance?

I guess, you had a big beat right now compared to where you thought you'd be, and so you did raise guidance.

I know there were some risks there that you mentioned.

But just how you think about that all.

On Series 6, I think, Mark, you were talking about basically -- I mean, like huge bookings, so I think better than everyone expected, were being sold out until 2020.

Now I think The Street could look at that and say, okay, they have nothing left to do.

So I guess how would you counter that?

What are we missing maybe on the margin side?

Or what's the next step?

Are you going to expand capacity more?

And sorry, I always see the C, but I'll leave it there.

Okay.

So Ben, I'll start with the guidance, and Mark would fill that in Series 6. So on the guidance side, we had some upside to the quarter from the settlement of the sales and use tax examination, which was in our suite of risks and opportunities for the year but not necessarily looked at for this being in Q1.

So that's an upside, and that's been offset partially by, as I mentioned, we pushed out the Beryl project.

So we're countering the sales process for that asset, and then we expect now to close that in 2019.

Based on the current position and the sales prices, we think we'll meet or potentially exceed our [site] expectations around that project.

However, as we've always said, we will look to optimize total project value by making a specific quarter or even a specific year of guidance, and right now, it's looking like we can optimize value on that asset by pushing it out.

We also had a couple of risks to the year.

So Mark talked a little bit about the potential impact of Series 6 around the aluminum on the frame.

We're also seeing -- see aluminum costs potentially increase across the EPC business as a function of tariffs, and then the ancillary impact to the pricing of exempted or nontariff steel and aluminum as well.

And then lastly, we have the Ishikawa project in Japan, which has had some schedule challenges related to (inaudible) weather.

And then we have a mitigation plan, and we still think we have a good path to selling that in 2018.

But if you look at all of those in totality, there's a lot of variability in the guidance for the year.

And we think that keeping the range where it is today gives us the flexibility should we see increase in tariff costs or potentially a slip-out of the Ishikawa project into next year.

So that's why we're keeping guidance for the year as it is.

In terms of the guidance over the year, we guided to A first half that could be about 1/4 of the earnings for the year.

If you look at the first quarter that we recorded to take the sales and use tax piece out and the breakeven guidance we're giving for Q2, you've got to about that same place.

So guidance for the year, not changed.

There are a lot of variability in the year, but we still -- we're pretty confident we have a good path to the bottom range of the guidance.

Yes.

And I'll add just on the guidance thing.

Ben, as you know, we're always going to be very prudent as it relates to when do we take up guidance.

And I think we're still early in the year, and there's a lot of moving pieces.

And we want to always reserve that optionality as it relates to capturing better value for development asset by selling it at the optimal time to capture that value equation versus trying to make it fit into a particular quarter or year.

So we're always going to be a little bit more balanced in that regard, and we'll see how the year progresses and provide an update around that in our next earnings call.

As it relates to Series 6, look, we're very happy with the bookings, and we've added more capacity now for that reason not only for what's contracts but also for what sits in our -- or excuse me, our mid- to late-stage pipeline at this point in time.

We're continuing to add EPC.

So we're taking and we showed another 380 megawatts or so this quarter that we booked on top of the module agreement that was already contracted.

We have noted there are 500 megawatts of EPC that we'll be able to most likely add to additional volume that we have contracted modules on, and we're going to continue to look through that.

So that's going to be another adder.

The other piece will be O&M.

On all the volume that we're talking about now goes to 1 gigawatt that we've added EPC to or will add EPC to here over the next quarter or so.

We're always working to add additional O&M to that scope, so another revenue stream and earnings for that will come on top of what we've already contracted on Series 6. The other piece of it is managing OpEx and scale.

And I think the thing we continue -- I want to make sure people have continued to model and think through is contribution margin flow-through against a relatively fixed cost OpEx.

So we now have a road map that says that, as we exit 2020, we'll be north of 7 gigawatts.

And [we're] modeling that and looking at incremental contribution margin again of fixed cost OpEx that's relatively flat to drive that margin expansion and EPS expansion as well.

So those are things I think that people need to look at.

And the other thing I want to make sure people will understand as it relates to the Series 6, we have a lot still in front of us.

We're happy with what -- where we started at this point in time, especially on the module wattage and efficiency.

I'm extremely pleased with that.

The risk that we all focused on the most was the front end of the line, and the front end of the line is performing extremely well.

We got some issues on the back end that we're trying to work through, and that's more of a throughput issue.

But the reality, though, is that as you think through the entitlement around the efficiency and ultimately, where the costs can go for Series 6, it's not done.

So I know Phil asked the question around 40% lower than where we were on Series 4. There's still meaningful opportunity to continue to drive down the cost on Series 6 and to continue to drive the efficiency up and further enhance the overall competitiveness of the product.

So those are all things that we'll give updates in the future.

But if you had to ask me maybe where people need to think about the business model as we move forward, those are probably the main points that I've mentioned.

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

Maybe, first off, Mark, you seem more constructive on topping the 1 gigawatt in the systems business bogey in recent quarters.

I know there's some moving pieces in terms of specific projects this year.

But given the recent trends and EPC traction you're seeing, is that 1 gigawatt bogey still the right target for 2018, maybe less 2018 but thinking about 2019 and beyond?

And then secondly, on the new Ohio plant, just a couple of modeling specifics, if you could help us out, Alex.

The $100 million for this year, fair to assume the other $300 million in CapEx comes in, in 2019.

And then at Analyst Day, I think you guys had said $50 million in startup and ramp penalty costs in each of '19 and 2020 related to capacity expansion.

How did that change with this new plant for those years?

I'll take the systems business question, Brian, and Alex can take the other 2. Yes, I'm very happy with where we are right now on the systems business.

We've always said that the gigawatt is the target annually, and we could trend above that number any given year.

And we've also said that maybe, over time, as we continue to see more progress, maybe that could end the number goes up.

And there's -- what I'm seeing right now, there's clearly a potential that, that could happen, especially with what we're seeing in some of the international opportunities beyond just what we're doing here in the U.S. So I think a combination of the 2 says that there could be an additional upside to the systems business as we look over the horizon, and we will be selective.

I mean, there is a very unique and specific opportunity set that we're going out after and where we want to position ourselves in the market as well as the geographies in which we want to do development in particular in the systems business in general.

But subject to kind of those guardrails, I'm very happy with the systems business momentum, and there could be an opportunity that we'll do better than that over the next few years.

And Brian, on the numbers, so just to clarify, it's $150 million of CapEx increase for this year and $250 million which we'll spend in 2019, so a total of approximately $400 million.

And on the startup side, we originally, at the Analyst Day, guided to $170 million of both ramp and startup combined in 2018 and about $50 million in 2019.

We're adding $10 million for that number for 2018 today.

So current guidance is $180 million combined.

Now We haven't updated for '19 or beyond yet, and we may do so later in the year.

But for now, the only update is $10 million addition to 2018.

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

Could you talk a little bit about how many new customers you booked business with this year?

How many opportunities you're seeing to actually bid in the products that are adding capacity on existing projects?

And then, is there any incentives to offset the incremental expense on the Ohio capacity expansion?

Yes.

So Colin, on new customers, we haven't historically reported on that specifically.

But what I can tell you, just -- I'll just use, for example, some of the announcements that we just made in terms of customers.

The Vectren, first deal that we've done with Vectren; Tampa Electric in which we announced the first deal a quarter or so ago.

First time we've done business with Vectren, the U.S. developer that we talked about 1.2 gigawatts that we've done 700 plus this year and 500 or so last year, that's essentially a new customer.

We really haven't done anything with that customer.

When we got into the international markets, we highlighted the volumes in Turkey, which, again, is a customer that we haven't done a lot of volumes with in the past.

Australia and some of the momentum that we got going on in Australia, we're expanding there.

Japan, we're expanding there.

So when I look at the profile of new customers, while we don't report on it specifically, one of the things that's enabled us to grow the pipeline that we have and in terms of contracted bookings as well as the mid- to late-stage, it's a very diverse customer set and a lot of customers that we have not done business with historically.

And part of this is because the excitement that's been created with Series 6, so the product that we have now is with that as competitively advantaged in the marketplace.

And there's even, to some extent, more of a pull versus a push with some of our customers wanting to come to us and to work with First Solar because of the technology that we're bringing to market from that standpoint.

I didn't understand.

I also talked about the incentive question, but you said something about adding capacity.

Help me understand that in terms of customer.

I thought I heard you say customers adding capacity.

I'm not sure of the question.

Yes.

There's existing sites where land has been unused where we have started to see a little bit of those [bookings] replacing modules or adding capacity around those existing sites within the existing interconnects.

So I'm just wondering if you've seen (inaudible) activity on that.

Yes, we have seen -- I mean, we haven't seen a lot of that.

We've seen -- we've actually done some of that in some of the international opportunities where we found ways to add incremental DC in certain cases.

Clearly, there's opportunities where customers have done development of a particular site and the interconnect has been sized to something that's larger than the actual capacity that was being built into the first phase and then gets expanded beyond that.

So those clearly do happen from time to time.

As it relates to any potential repowering, at least from a First Solar perspective, we haven't seen much of that at all.

Yes, Colin, on the Ohio side, so yes, so the range of state and country level local incentives, we haven't listed exactly what those are.

But they cleared a range of job creation incentives, training incentives, sales and property tax abatements, most of which have been largely finalized but have potentially got either local (inaudible) I think completely finalized.

So they're largely in place, and there's a wide range.

The other thing I'd say is that the -- one of the reasons we can expand U.S. manufacturing capacity is associated with a lot of the lower corporate tax rate and tax reform.

And part of that tax reform has been the ability to immediately expand sales on equipment, so less a direct incentive but also another reason this may be a small competitor relative to international manufacturing.

Yes.

What I said in my comments as well, I mean, the relative competitiveness of U.S. manufacturing right now is really exciting to me and really, to some extent, overwhelmingly in terms of this relative position.

It's not just the corporate tax rate and the incentives and everything else that have been brought to the table but just the labor productivity and advantages that we can get here in the U.S. relative to what we can see in some of the international markets and just the general support we can get at the state level and the federal level support, whether tax structure or whatever it may be.

U.S. manufacturing, coupled with our Series 6 product that less labor content per module, it's very competitive to the U.S. manufacturing right now, and it also helps us reduce the logistics, the freight costs, delivering to customers here in the U.S., our ability to serve our customers more timely and responsiveness around that.

We're very happy about the ability to add more manufacturing capacity in the U.S.

And we'll take our next question from Michael Weinstein with Crédit Suisse.

Could you talk about the state and local incentives and when you think they might start to flow in for the new factory?

Yes.

So Mike, there's not much update I can give you beyond the range of those when our price is being finalized.

But we'll update you further on a later call as we get more clarity.

Yes.

And one thing, I mean, I would just say, to some extent, some of the incentives that we're getting are comparable to the incentives we would've had in -- when we first started manufacturing in Ohio.

And a lot of them will relate to property tax abatements and other reductions that'll happen over time.

So the numbers, when we're actually able to communicate them, there'll be a large number but -- in aggregate, but you also have to look at it at the time line of which you'll see the benefits will carry on for many years.

And we'll take our last question from Edwin Mok with Needham and Company.

So I guess, 2 last question.

First, on the U.S. capacity.

I think you talked about higher automation for Series 6 and, therefore, lower labor costs.

But then, I would suspect there might be some higher -- potential other things that might be higher cost, yes?

Is that what you're thinking about?

Is it more costly to produce here?

Or is it less costly to produce here?

And then having that additional capacity in the U.S., does it help you be more aggressive from bidding for systems project?

And just one quick question on OpEx.

If you take up your startup cost by $10 million, that means it's actually lower your OpEx for the year, right, excluding startup costs, is that correct?

And kind of to Mark's point about leverage in the model, long term, is this like $283 million OpEx a level that you think you can sustain even as you ramp to that 7 gigawatt comp capacity that you eventually will be selling?

Yes.

So I'll let Alex take a handful, and I'll just take kind of the first one as it relates to the competitiveness of U.S. manufacturing.

Yes, Series 6 is lower labor content per module produced.

So that helps relative -- and again, you have better labor productivity in the U.S. So a combination of less labor content and then better productivity.

In fact, it creates almost a level playing field with our manufacturing in Vietnam and in Malaysia.

The other thing that we get, though, is localization of the supply chain.

And so having a local supply chain now that we have higher production volumes that we can run through the supply chain, which drives the economies of scale to our suppliers, which enables them to further reduce their costs, which drives flow-through benefit to First Solar is extremely impactful as well.

So -- and you couple that with lower freight costs to deliver to customers.

And the reality is that, yes, while we have a tax holiday in Malaysia, when you look through that holiday period and you look at the normalized U.S. tax rate relative to the normalized Malaysia or Vietnam tax rate, it's favorable from that standpoint.

So there's many different aspects that has enhanced the overall competitiveness of U.S. manufacturing.

And like I said, we're very happy that we'll able to expand manufacturing in Ohio.

And I'd say on the -- for Series 6 in Ohio directly impacts (inaudible) bid on systems projects.

I don't know that having more U.S. manufacturing relative to other international manufacturing changes our ability to be on the systems side.

I would say, overall, Series 6 is a very competitive product, and having that product enables us to be more competitive in the systems business, be it domestically or internationally.

But one of the key things -- you talked about OpEx and scale, so yes, we're going to be up about $10 million of startup this year, which does mean we are managing our core SG&A effectively.

That will be down slightly from previous guidance.

This year, we're going to be running still at around $0.10 per watt of OpEx.

And if you look at our growth over time, we aim to grow the company on 7-plus gigawatts of capacity without significant increase in OpEx.

There'll minimal increases around some of our variable costs, including sales expense.

But outside of that, we would look to bring that number down by 50% or more to get down to, call it, a $0.05 per watt of OpEx number out when where at that comes 7 gigawatts range.

And also, remember that when we talk about that OpEx number, we're including freight and warranty in that number, which is not the same if you look at most of our competitors.

So it's still apples-to-apples comparison.

You have to strip that out of our number.

So as Mark mentioned, it's clear value in scaling and the benefits of OpEx in the incremental contribution margin that we can get through managing our OpEx profile carefully.

And this does conclude the question-and-answer session and the First Solar's First Quarter 2018 Earnings Call.

We thank you all for your participation, and you may now disconnect.