U.S. Markets open in 1 hr 37 mins

Edited Transcript of FSLR earnings conference call or presentation 1-Aug-19 8:30pm GMT

Q2 2019 First Solar Inc Earnings Call

TEMPE Aug 4, 2019 (Thomson StreetEvents) -- Edited Transcript of First Solar Inc earnings conference call or presentation Thursday, August 1, 2019 at 8:30:00pm GMT

TEXT version of Transcript

================================================================================

Corporate Participants

================================================================================

* Adrianna Defranco

First Solar, Inc. - Financial Analyst of Financial Planning & Analysis

* Alexander R. Bradley

First Solar, Inc. - CFO

* Mark R. Widmar

First Solar, Inc. - CEO & Director

================================================================================

Conference Call Participants

================================================================================

* Brian K. Lee

Goldman Sachs Group Inc., Research Division - VP & Senior Clean Energy Analyst

* Colin William Rusch

Oppenheimer & Co. Inc., Research Division - MD and Senior Analyst

* Julien Patrick Dumoulin-Smith

BofA Merrill Lynch, Research Division - Director and Head of the US Power, Utilities & Alternative Energy Equity Research

* Michael Weinstein

Crédit Suisse AG, Research Division - United States Utilities Analyst

* Philip Shen

Roth Capital Partners, LLC, Research Division - MD & Senior Research Analyst

* Travis Miller

Morningstar Inc., Research Division - Director of Utilities Research and Strategist

================================================================================

Presentation

--------------------------------------------------------------------------------

Operator [1]

--------------------------------------------------------------------------------

Good afternoon, everyone, and welcome to First Solar's Second Quarter 2019 Earnings Call. This call is being webcast live on the Investors section of 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 Adrianna Defranco from First Solar Investor Relations. Ms. Defranco, you may begin.

--------------------------------------------------------------------------------

Adrianna Defranco, First Solar, Inc. - Financial Analyst of Financial Planning & Analysis [2]

--------------------------------------------------------------------------------

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

Today, the company issued a press release announcing its second quarter 2019 financial results. A copy of the press release and associated presentation are available on 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 begin by providing a business and technology update. Alex will then discuss our financial results for the quarter and provide updated guidance for 2019. Following their remarks, we will open the call for questions.

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?

--------------------------------------------------------------------------------

Mark R. Widmar, First Solar, Inc. - CEO & Director [3]

--------------------------------------------------------------------------------

Thanks, Adrianna. Good afternoon, and thank you for joining us today.

I'll begin by noting our second quarter loss of $0.18 per share. Despite this result, due largely to increased variable compensation expense associated with the company's short-term and long-term incentive plans and increased tax expense, which Alex will address, we've put a number of wins on the board during the quarter and are maintaining our full year earnings per share guidance.

The Series 6 program remains on track, with improvement across all key manufacturing metrics. Q2 saw the greatest quarterly production run in the company's history.

From a bookings perspective, we recorded our largest-ever individual order for Series 6 modules. And from a shipment perspective, we recorded record quarterly shipments. These highlights our reflection of the team's continued excellence in a very dynamic and changing business environment.

Turning to the market. Catalysts driving increased PV penetration continued to point to growing global momentum and strong demand. On our last earnings call, we discussed our optimism for utility-scale solar growth driven not just by pure economics but by the groundswell of communities making strong actionable commitments to renewable energy. The recent uptick in government's commitments, coupled with growth and corporate activity, provide the underpinnings for secure and stable growth.

In many regions, the solar industry has reached a cost inflection point relative to coal. In the U.S., the Energy Information Administration has identified a trend of younger and larger coal plants shutting down. And a study by Energy Innovations released in March showed that it would be cheaper to replace 74% of U.S. coal with new wind and solar. The study further found that replacing 94 gigawatts of existing coal plants with wind and/or solar would result in a 25% reduction in energy cost.

In Europe, the first half of 2019 saw renewable energy outproduced fossil fuels for the first time. These are just some of the encouraging signs of sustainable growth we see.

Shifting to the domestic market. We've pointed out during our last earnings call that 3 states have adopted active mandates to reach 100% clean electricity standards, with over a dozen or more committed to nonbinding goals. As of today, 8 states enacted 100% clean and renewable energy goals and an additional 29 states, plus Washington, D.C., are targeting 9 binding energy targets.

Another trend gaining traction in the U.S. relates to battery storage. Deployment of grid-connected energy storage in the U.S. are expected to hit over 700 megawatts this year and are projected to reach 2.5 gigawatts by 2023. Moreover, the economics of these projects signal that, in certain regions today, clean and dispatchable energy can be made available for less than the cost of new natural gas.

We also continue to see growing momentum along corporates seeking to decarbonize their electricity. 2018 saw over 13 gigawatts of corporate PPA agreements, doubling 2017 levels, with new buyers and emerging markets tipping the scale.

Since our comments last quarter, companies in France and Poland signed corporate power purchase agreement for large-scale solar, making a first for these nations.

We continue to solidify our position as a leading global provider of corporate solar solutions by providing clean or eco-efficient solar technology. On Tuesday this week, we just announced jointly with Microsoft that our Sun Streams 2 solar facility will power Microsoft's new energy-efficient data center being built in Arizona. We are thrilled Microsoft values our Series 6 technology, especially given its lower carbon footprint and superior environmental profile compared to crystalline silicon. We look forward to providing Microsoft cleaner solar electricity.

Earlier in the quarter, we announced that our Cove Mountain 2 solar power plant would support Facebook's Eagle Mountain data center in Utah through a PPA with Rocky Mountain Power. The project will be constructed near the 58-megawatt Cove Mountain power plant, which we announced last year, will also support Facebook's operations.

Continuing this trend, we announced last week that Kellogg's Australia and New Zealand signed a PPA with the Beryl Solar Farm developed and operated by First Solar in New South Wales.

Finally, the utility scale market in the United States continues to thrive. Notably, we're seeing an increase in multiyear module sales agreements driven by our customers' need for pricing and technology certainty and our commitment to stand behind our contracts.

Starting on Slide 4, I'll provide an update on our Series 6 production metrics. As a reminder, we began production at our first Series 6 factory in April 2018. And over the subsequent 5 quarters, that continued to ramp production in the U.S., Malaysia and Vietnam. Leveraging our controlled replication process, we are operating 4 factories, which are matched in terms of key processes and which have produced 1.9 gigawatts year-to-date.

Reflecting on the significant progress we've made over a relatively short period of time, we are pleased with the Series 6 in terms of schedule, performance and cost.

On a fleet-wide basis, since April, we have seen significant operational improvements. When comparing performance for the month of April to July, meaningful improvement can be seen. Megawatts produced per day is up 16%. Capacity utilization has increased 12 percentage points to 94%. Adjusted for plant downtime, the fleet capacity utilization was 96%. Note, in support of our module efficiency road map in July, we increased the volume of engineering test articles, which adversely impacted capacity utilization by 1 percentage point. The production yield is up 2 percentage points to 91%. The average watt per module has increased 3 watts. And finally, the percentage of modules produced with antireflective coating has increased by 5 percentage points to 91%. The combination of our efficiency improvement program and increased ARC utilization has led to a significant improvement in the module bin distribution. In July, the ARC bin distribution between 420-watt through 430-watt modules represented 87% of production. Our best factory had ARC utilization of 95%, with 94% of the production volume at a 420-watt bin or higher. With 893 megawatts produced, Q2 continues our planned production ramp, and we have carried the momentum into July where we are experiencing our best-ever production month with 322 megawatts produced.

On an individual plant level, all of the factories are performing well. At our second factory in Vietnam, the production metrics are ahead of plan and within 6 months of commencing operation, consistent with the fleet average. This is a direct result of learnings from prior factory ramps and the synergies realized by having a multi-factory site, which effectively leverages the WIP inventory buffers across 2 factories and creates additional equipment redundancy. We would expect to see similar benefits when we started up our second factory in Ohio, where construction is continuing according to schedule, the first tools installed in June.

Note as a highlight during our Q1 earnings call, our new Ohio factory will include additional capital, which among other improvements will target a 5% increase in the nameplate capacity. Depending on our ability to realize the targeted capacity increase, we will look to roll out this program across the fleet. We anticipate starting the production in early 2020 with the possibility of prequalified production in Q4 2019.

The progress we have made ramping our factories has been a key contributor, enabling the achievement of our Series 6 cost per watt objectives for the first half of 2019. While this is a significant accomplishment, there is tremendous amount of work still in front of us in order to achieve our cost per watt road map for the full year. As we noted in our April's earnings call, our Series 6 cost per watt is expected to drop approximately 30% from Q1 to Q4.

Turning to Slide 5, I'll next discuss our bookings activity. In total, we have 4.3 gigawatts of net bookings in 2019, including net bookings of over 2 gigawatts since the last earnings call. After accounting for shipments of 2.2 gigawatts in the first half of the year, including record quarterly shipments of 1.4 gigawatts in the second quarter, our future expected shipments are 12.9 gigawatts.

As it relates to the systems business, we converted 2 opportunities in Japan from our mid- to late-stage pipeline into 77 megawatts of bookings, with expected deliveries through 2022. With these projects, our footprint in Japan has grown to approximately 400 megawatts. We continue to believe in the strength of our portfolio of systems opportunities in Japan, and we could double our existing systems bookings there by the end of 2023.

The remainder of our net bookings for the quarter were module-only bookings essentially, all for Series 6 product. International markets represented slightly more than 100 megawatts of the bookings. As I noted previously, we are seeing an increase in multiyear module sales agreements driven by our customers' need for certainty in terms of the technology they are investing in and the certainty that we will stand behind our contracts. Representative of this, we have secured our largest Series 6 agreement with a new customer to supply 1.7 gigawatts of deployment of projects across the U.S. We've also secured a 0.3-gigawatt booking with another new customer in the United States. Note, this is the first phase of the project with a similar size second phase included in our contracted pipeline and awaiting confirmation of conditions precedent to become a booking.

We are particularly pleased with the strength of the bookings in the quarter despite the decision in June of the office of the U.S. Trade Representative to exempt bifacial panels from Section 201 tariffs. While we are able to contract through this headwind, it's important to note that the disappointing decision, which in our view has the effect of undermining the administration's efforts to secure a level playing field for U.S. solar manufacturing, introduces a new source of uncertainty going forward.

Since the last earnings call, related to certain customer and specific events, we debooked 0.3 gigawatts scheduled to be shipped in 2019 and 2020. These events included the combination of a customer's request to support their revised project development schedule and project size as well as settlement of an ongoing dispute with a customer, which originated over the potential sale of one of our systems projects. Although these events had an adverse impact to our previously contracted module backlog, given our robust pipeline, we were able to contract the volume with other customers. More importantly, we view the resolution as an investment in long-term customer relationship.

Turning to Slide 6. We show the forecasted supply plan for the second half of 2019 through full year 2021. As noted in the Q1 earnings call, we are effectively sold out through the remainder of 2019 and full year 2020. With the most recent bookings, 2021 is starting to book up relative to an anticipated supply plan of 6.5 gigawatts. For full year 2021, we booked and contracted volumes subject to conditions precedent representing approximately 60% of the supply plan. This leaves approximately 2.5 gigawatts to be booked in 2021. Note, approximately half of the remaining volume is anticipated to be needed for our self-development systems business. Assuming our systems business demand is level-loaded by quarter in 2021, we are largely sold out for third-party module sales in the first half of 2021.

We remain pleased with the bookings momentum and have increased confidence in achieving or exceeding our target 1:1 book-to-ship ratio in 2019 even as we continue to contract for deliveries over 2 years in advance.

Slide 7 provides an updated view of our mid- to late-stage bookings opportunities, which now totals 6 gigawatts DC, a decrease of 0.6 gigawatts from the prior quarter. When factoring the bookings for the quarter, 0.6 gigawatts, which were included in the opportunities in the prior quarter, our mid- to late-stage pipeline remains unchanged. Additionally, the pipeline includes 0.6 gigawatts of confirmed opportunities awaiting satisfaction of outstanding conditions precedents before being booked in the quarter.

As a reminder, our mid- to late-stage pipeline is reflective of those opportunities we could feel would book within the next 12 months and is a subset of a much larger pipeline of opportunities, which totals 13.3 gigawatts DC, which increased 1.6 gigawatts from last quarter. This includes 0.8 gigawatts of opportunities in 2019 and 2020, which provide demand resiliency to our near-term bookings production, while the remaining 12.5 gigawatts of demand would be for modules delivered in 2021 and beyond.

In terms of geographic breakdown of our mid- to late-stage pipeline, North America remains the region with the largest number of opportunities at 4 gigawatts DC. Europe represents 1.5 gigawatts, with the remainder in Asia Pacific.

In terms of segment mix. Our mid- to late-stage pipeline includes 1.9 gigawatts of systems opportunities across U.S. and Japan, with the remainder being module-only sales. The significant increase in systems opportunities is largely attributed to U.S. opportunities associated with the ITC Safe Harbor.

We have also had significant success across our energy services business, with year-to-date bookings of over 1.6 gigawatts DC. Approximately 0.5 gigawatts was associated with the sale of development assets, with the remainder 1 gigawatts was contracted with projects developed by third parties. This brings our total energy services portfolio of assets on the contract to nearly 13 gigawatts globally.

Before turning the call over to Alex, I'd like to discuss recent developments related to the ongoing class action lawsuit, which was filed in 2012. As previously disclosed, in August of 2018, we filed a cert petition with the U.S. Supreme Court concerning the appropriate standard to determine loss causation. We identified various standards across several circuits and did not believe the standard used in the Ninth Circuit, which is a standard that will apply to this case, was appropriate. We were supported by the U.S. Chamber of Commerce, the Securities Industry and Financial Markets Association, the Business Roundtable and other groups and filed an amicus brief with the court in support of our position. At the end of last month, the Supreme Court denied the cert petition. While we are disappointed with the denial, we continue to believe we have meritorious defenses and are vigorously defending this case. Also, as previously disclosed from following the results of the Supreme Court, the Arizona District Court ordered that the trial begin in January 2020.

I'll now turn the call over to Alex who will discuss our second quarter financial results and provide updated guidance for 2019.

--------------------------------------------------------------------------------

Alexander R. Bradley, First Solar, Inc. - CFO [4]

--------------------------------------------------------------------------------

Thanks, Mark. Before discussing the financial results, of course, I'd like to note that as we say from our previous earnings call, we anticipated that Q2 would be a breakeven to last quarter, and our second quarter financial performance was in line with those expectations. With the sale of the Beryl asset in Australia and the Cove Mountain and Muscle Shoals assets in the U.S., continued progress towards additional system sales and an increase in third-party module sales during the quarter, we remain on track to hit our financial objectives for the year.

With regards to the sale of the Cove Mountain and Muscle Shoals assets, these projects are not anticipated to begin construction until late 2019 and early 2020, respectively. However, as noted on the previous earnings call, given the opportunity to optimize valuations and reduce risk, we sold these assets prior to notice to proceed.

In addition, in lieu of our traditional sales structure whereby we would perform an EPC and recognize revenue on a percentage of completion basis over the life of the contract, we have, in these cases, sold the project companies with a module sale agreement, and our customers will engage a third-party EPC provider. This structure has several advantages, including earlier cash collection and reduction of EPC risk exposure as we continue to evaluate the extent of our offering of this service.

Note that in our public filings, this also has the effect of removing the sold assets from our systems pipeline table, adding approximately 0.5 gigawatt of volume into the module backlog. With respect to the module backlog that will be updated in our forthcoming 10-Q, although this metric is always impacted by rounding given it is reported in gigawatts and billions of dollars, this quarter is also atypical circumstances which will imply a lower ASP for new bookings within the case. This is a result of 2 key drivers: firstly, with regards to the aforementioned Muscle Shoals and Cove Mountain projects, a change in the sales structure results in a great upfront margin and the approximately 0.5 gigawatt of additional module backlog with deliveries in 2020 and 2021 comes at ASPs lower than the previous average. This has not changed the overall project economics for these assets but has the impact of lowering the overall average ASP and the module backlog. Secondly, as Mark noted, we had 0.3 gigawatts of debookings with a certain customer since the last earnings call, which resulted in an ASP reduction associated with those volumes. Combined, these changes lower the overall average ASP in the backlog. The implied average ASP for the calendar quarter is not reflective of the pricing associated with new bookings, including the 2 gigawatts of net bookings since the prior earnings call. And to be clear, we're very pleased with our overall ASPs and new bookings in 2019.

Turning to Slide 9. I'll start by covering the income statement highlights for the second quarter. Net sales in Q2 were $585 million, an increase of $53 million compared to the prior quarter. The increase in net sales is primarily a result of increased module sales as well as the closing of the sale for fully constructed Beryl project in Australia and the sale of our Cove Mountain and Muscle Shoals projects in the U.S. Although Q2 is a record shipment quarter for us, given the contractual terms associated with certain third-party module bookings as well as shipments to our systems projects, which have not been sold, we only recognized revenue on approximately 1/3 of those shipments.

On a segment basis, as a percentage of total quarterly net sales, our systems revenue in Q2 is 61% compared to 63% in Q1.

Gross margin was 13% in Q2 compared to breakeven in Q1. The systems sale gross margin was 18% in the second quarter, and the module segment gross margin was 5%. As a reminder, module segment cost of sales is comprised of all third-party module cost of sale as well as Series 6 ramp-related costs. We experienced ramp-related costs of $18 million in the second quarter, which combined with Q1, is equivalent to the majority of the forecasted ramp charges for the year.

Regards to the module segment, gross margin increased 18 percentage points from negative 13 in Q1 to positive 5 in Q2. This is achieved through a combination of increased shipments driven by increased Series 6 volume, lower ramp costs and lower cost per watt.

Operating expenses were $86 million in the second quarter, an increase of $9 million compared to Q1. This was driven by increased variable compensation associated with the company's short-term and long-term incentive plans.

We had an operating loss of $9 million in the second quarter compared to an operating loss of $77 million in the prior quarter. This was a result of the increased module and systems revenue and margin referenced earlier, partially offset by the increase in operating expenses.

We took a mark-to-market charge of $4 million related to the fair value of certain interest rate swap contracts for some of our project assets in Japan and Australia. This is a timing impact based on movement of interest rates within the quarter, approximately half of which was recovered during the quarter through the increased sale value of the Beryl project. We expect the remainder to be recovered in subsequent quarters through increased project values recorded at the time of sale.

We recorded tax expense of $12 million compared to tax benefit of $1 million in Q1. The increase in tax expense for the quarter is attributable to the jurisdictional mix of income as well as a discrete return to provision expense of $7 million. Combination of the aforementioned items led to a second quarter loss per share of $0.18 compared to a loss per share of $0.64 in the first quarter.

I'd like to turn to Slide 10 to discuss select balance sheet items and summary cash flow information. Our cash and marketable securities balance ended the quarter at $2.1 billion, a decrease of approximately $170 million from the prior quarter. Total debt at the end of the second quarter was $481 million compared to $571 million at the end of Q1. The reduction in debt is primarily due to the sale of the Beryl asset in Australia and the assumption by the buyer of the project-level debt. As a reminder, all of our outstanding debt continues to be project-related and will come off our balance sheet when projects are sold.

Our net cash position decreased by approximately $80 million to $1.7 billion. The decrease in our net cash balance is primarily related to capital investments in Series 6 manufacturing capacity.

Net working capital in Q2, which includes noncurrent project assets and excludes cash and marketable securities, decreased by $133 million versus the prior quarter. Change was primarily due to collections on receivables for systems projects, partially offset by an increase in our Series 6 module inventories. Cash flows from operations were $13 million in the second quarter.

And finally, capital expenditures were $179 million in the second quarter compared to $149 million in the first quarter as we continued Series 6 capacity expansion.

Continuing on to Slide 11, I'll next discuss the updated assumptions associated with our 2019 guidance. Firstly, our guidance continues to assume both a back-ended Series 6 module sale profile as well as a significant Series 6 cost reduction rig profile over the next 2 quarters. As Mark noted earlier, our Series 6 cost per watt is expected to drop approximately 30% from Q1 to Q4. We also continue to expect approximately 60% of our ramp-related and startup charges to have been occurred in the first half of the year.

Secondly, while we continue to forecast the closing of multiple project sales, both in the U.S. and internationally in the second half of the year, timing between Q3 and Q4 remains uncertain and could have a material impact on the timing of revenue and earnings between the third and fourth quarters.

In regards to our U.S. project sales, we continue to be pleased with the progress made in the sales processes. But over the assets with offtake agreements with FCE, recent legislative developments in California have been positive while uncertainty around timing and value remains. As indicated on the previous call, should the market not reflect what we believe to be their appropriate risk profiles and values for the assets, we would look to refinance the assets and hold them on balance sheet through the period of uncertainty rather than selling at prices below what we believe to be fair value.

As a reminder, whilst unlikely should this occur, it could result in full year EPS approximately $0.50 over the low end of the guidance range.

Thirdly, we continue to highlight that our guidance does not reflect the potential for higher legal costs associated with defending the class action lawsuit, but these defense costs may exceed our insurance coverage limits. In addition, our guidance does not take into account the financial impact of any resolution of that lawsuit given the uncertainties of trial.

With these factors in mind, we're updating our 2019 guidance as follows: net sales guidance remains unchanged. Gross margin guidance has increased 50 basis points to a revised range of 18.5% to 19.5% due to higher upfront recognition of margin on the Muscle Shoals and Cove Mountain projects associated with the sales structure discussed earlier, offset by an increase in Series 6 ramp-related costs. Operating expense forecast has been lowered by $10 million to a revised range of $360 million to $380 million as a result of decreased plant startup expense, which is now forecast to be $55 million to $65 million, partially offset by higher variable compensation expense. The operating income guidance has been increased to a revised range of $290 million to $340 million as a result of the above changes.

Full year tax expense is now forecast to be approximately $70 million, up from our previous estimate of $50 million. The increase is primarily due to a shift in jurisdictional mix of income and the previously mentioned return to provision expense. And earnings per share, net cash, capital expenditures and shipment guidance is unchanged.

Finally, I'll summarize the key messages from our call on Slide 12. Firstly, we continue to be pleased with the progress on our Series 6 platform, including the significant progress across key manufacturing metrics. Secondly, our year-to-date bookings of 4.3 gigawatts continue to strengthen our future contracted position, and we're now approximately 60% contracted in relation to our 2021 supply plan. And lastly, we have increased our gross margin operating income guidance and maintained our full year EPS guidance range.

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

================================================================================

Questions and Answers

--------------------------------------------------------------------------------

Operator [1]

--------------------------------------------------------------------------------

(Operator Instructions) Your first question comes from Philip Shen with Roth Capital Partners.

--------------------------------------------------------------------------------

Philip Shen, Roth Capital Partners, LLC, Research Division - MD & Senior Research Analyst [2]

--------------------------------------------------------------------------------

I have three. The first one is recall the slide that you guys put out back in Q4 of '18. I think it's Slide 12 where you had your average cost per watt. And Q1 was 30% above that; Q2, 5% above that; and then Q4 of '19 was expected to be 10% below average for the year. Can you guys just give us an update as to where things stand relative to that? Mark, I know that you talked about meeting your objectives and so forth. How does things stand relative to that? First. And then secondarily, as it relates to the bookings, how are bookings looking into 2022 with this bifacial exemption? Are customers starting to look more closely to book bifacial offerings by your peers? And then finally, can you give us an update on your view of -- your latest view on capacity expansion and, in general, maybe shareholder capital -- capital allocation in general?

--------------------------------------------------------------------------------

Mark R. Widmar, First Solar, Inc. - CEO & Director [3]

--------------------------------------------------------------------------------

All right, Phil, on the cost per watt, and -- what I did say in my prepared remarks was that we're happy with where we are right now. We met our first half commitment on the reduction. And as you remember, we had a pretty steep reduction from first quarter into the second quarter, so we're happy with the first half results. Still work to be done yet to get to the second half. A number of things. If you look at the first half, I mean, a lot of it was driven by -- utilization has come up significantly. We've driven up throughput improvements dramatically. Yields have improved significantly. They're all in the metrics that we reflected in the deck and I referenced it in my comments. And you can clearly see the benefit of that, it will drive down the cost per watt. As I look across what still needs to happen for the balance of the year to get to our cost per watt, it's -- there's no one silver bullet. It's one of many things. We've got labor that needs to come out. When we do the ramp process and starting up production, we threw a lot of labor into the manufacturing process that we're now in the process of taking out. So there's a pretty significant reduction of labor that should get down to the entitlement where we expect it to be. But in the early ramp phases, we threw labor into the process in order to continue to run operations at the highest throughput levels that we can. We've got some building material opportunity still that we've got to capture. We've got negotiations ongoing right now that -- to realize that benefit. And then on top of that, there's still -- we've got to get the yields up. Yields are at 91%. We've got to get those numbers up past 95%, closer to 97%, so there's some work to be done there. We still have a little bit of room to go on utilization. So we've got to see that and continue to drive throughput as -- to the highest possible optimization that we can. So there's a lot of levers, not one individual lever. But we're still confident with what we still need to do and our ability to get there. But a lot of the heavy lifting still to go. We have to get to the cost per watt for the second half of the year.

Bookings. We're very happy with bookings and where we are. And we can see from the slide that the bifacial exemption happened, I think, it was either June 12, or something along that line, the bookings that we were able to secure. We had a couple of 2 key new customers all -- both happened after that date. Not to say that there was an inflection point in discussions with the customer around ramifications around bifacial modules. I don't have any concern around our ability to compete with a bifacial module. It's just a discussion around where the economics lands. So to me, the bigger challenge we have if the exemption stays where it is, that it ultimately compromises economics and the realization of what we can capture relative to our competition. The spirit of what was put in place under the 201 was to allow companies to scale and to ramp manufacturing and to enhance overall cost competitiveness to deal with imported competitors. We made an investment -- a significant investment in our factory in Ohio for that very reason, coupled with the Tax Reform that was provided. And what we had hoped for would be allow our ability to ramp our new facility in an environment that wouldn't be under siege by a flood of imports. And I do think there is risk that, that could happen with the exemption as it's currently written, unless there's some form of a modification to it or either recision or some form of quota. I think it will create some challenges that we'll have to be able to deal with. So we're very disappointed in that decision, as I said in my comments.

Capacity expansion. If you look at 2021, yes, we highlighted in the slide, the committed supply plan is basically about 6.5 gigawatts. If you look at it, coupled with our systems business that we have, we anticipate to have in 2021, we have about 1, 1.25 kind of available capacity still in 2021. And we're focused on what we can do to potentially address that. We're very encouraged to hopefully see positive results in what we're doing in Perrysburg. As I indicated in our prepared remarks that if -- that plant will go into the ground with additional capacity, about 5% to 7% incremental capacity, which we'd say that at, a nameplate basis, we could drive somewhere in the range of 60 to 80 megawatts, somewhere in that range of incremental throughput if we're successful. And what we'd like to be able to do is to then replicate that across the balance of the fleet and continue to look for other opportunities for throughput. If you look at my prepared remarks, one of the things we highlighted was that the benefit we've seen in Vietnam is to have 2 factories co-located at 1 site, whether it's the leveraging of the WIP buffers, whether it's the redundancy that you have around the tool set across both factories. So we know we want to leverage that across Malaysia, Vietnam and the U.S. We've got the U.S. covered now, Vietnam is already addressed and then, we're in the process, but end of this year, we'll start the conversion of Malaysia. But then beyond that, we just want to continue to drive throughput because 2 things: efficiency and throughput are going to drive to our lowest cost per watt, and so we're very focused on doing that. We're continuing to evaluate options around how we can get additional capacity out of those factories. But we still have the optionality if we choose to convert KLM 1 2. And so that's another 1.2 gigawatts that could be put into the mix. My first choice would be to optimize the efficiency and the throughput leverage first and then move to an expansion or, I should say, a conversion to KLM 1 2. A lot of different options and things and scenarios that we'll have to evaluate, but very happy with the demand and supply backlog that we currently have and contracted that really almost through the next 10 quarters.

--------------------------------------------------------------------------------

Operator [4]

--------------------------------------------------------------------------------

Your next question comes from Julien Dumoulin-Smith with Bank of America Merrill Lynch.

--------------------------------------------------------------------------------

Julien Patrick Dumoulin-Smith, BofA Merrill Lynch, Research Division - Director and Head of the US Power, Utilities & Alternative Energy Equity Research [5]

--------------------------------------------------------------------------------

Maybe to just pick up where you just left off. In fact, why don't we talk a little bit, if you can, on this cost per watt trajectory, not just heading into 4Q? And I appreciate what a lift that may be already, how do you think about that over time here, especially as you think about scaling the megawatts here, too? I mean, just -- I want to focus on the cost per watt metric and where we can go from here. And I don't want to preempt too much guidance into '20 specifically, but trajectory-wise, how much more is there to go, sort of, overall, especially as you see these other products continue to have trajectory on cost themselves, be it bifacial or otherwise?

--------------------------------------------------------------------------------

Mark R. Widmar, First Solar, Inc. - CEO & Director [6]

--------------------------------------------------------------------------------

Yes. I mean, it's just hard to go forward with too much details and specifics. So -- but what we said before, even with the commitment when we first launched Series 6 and then obviously, a significant cost reduction from Series 4, but that initial indication, which was about a 40% reduction for relative to our Series 4 price -- or cost, excuse me, but that wasn't necessarily an endpoint of a destination that we still have room to run. And the 2 most significant levers around that are continuing to drive up our efficiency. And then if we can drive incremental throughput across each of our factories, those will drive -- have some significant impacts on the cost per watt and enable us to continue to drive the cost down to create an advantage position on a cost position to our competition, which is the ultimate goal that we have. Clearly, it is challenging in terms of what we're seeing in the marketplace with some of our competitors and their cost, but then we think the final destination still will allow us to have an advantage position. There's other things we got to work through though that helps us get to that destination. One is the frame. We've talked before about the frame and even the design of the frame is actually adding cost to the module. So we've got to optimize against the frame. And there really is an optimization across the frame and the glass. So it's a combination of glass thickness and the associated frame, the packaging. Those 3 components, the 2 sheets of glass and the frame make up a very significant percentage of the overall build material. So how do we optimize against that and drive that down from where it is right now. And we've got a number of options around how to do that. And then we just need to continue to drive and improve our yields. So I would say the other one that's a little bit of a headwind right now is our yields. As I highlighted, we're at 91, and we need to drive that up. So it's not only driving the yields up. It's also where does scrap occur within the manufacturing process. And the more that we see the scrap loss in the back end of the factory, it just drives a higher cost because majority of the volume is already incurred at that point in time. So there's a lot of things that we need to do along those lines. But I don't think there should be a view that there's a -- there's not an additional rooms still to go and continue to drive towards the lowest optimal cost point. A lot of work though to make it happen, but we got to first figure out in the second half of this year and then carry it forward from there.

--------------------------------------------------------------------------------

Julien Patrick Dumoulin-Smith, BofA Merrill Lynch, Research Division - Director and Head of the US Power, Utilities & Alternative Energy Equity Research [7]

--------------------------------------------------------------------------------

If I can just quickly ask on the incremental bookings for ASP. How firm is the market at this point? We saw some uptick in pricing for the first time in a while in the U.S. Obviously, bifacial is impacting things now. But just where are we on the incremental bookings. And just if you can talk a little bit on market pricing today as you've already alluded to for bifacial.

--------------------------------------------------------------------------------

Mark R. Widmar, First Solar, Inc. - CEO & Director [8]

--------------------------------------------------------------------------------

Yes. So look, I'm really happy with the 2 gigawatts or so -- north of 2 gigawatts on a gross basis of bookings. I mean the ASPs that we're realizing on the incremental volumes is consistent with what we saw in the first half. So if I look across the north of 4 gigawatts of bookings that we have year-to-date, that ASP is very firm and consistent really across that entire volume. And the really only variation you have is what year are you shipping it, right? So some of the bookings that we had in the first quarter still were 2020. Now we're all into '21. We've got some bookings out in '22 and even some that touch into '23. So the real variation that you see right now around ASP depends on the year which you're booking against. But we've been pretty, pretty pleased with how firm the market has been.

--------------------------------------------------------------------------------

Alexander R. Bradley, First Solar, Inc. - CFO [9]

--------------------------------------------------------------------------------

Yes. Julien, the one thing we talked about in the prepared remarks I want to make clear is the difference between what you're going to see in the Q when it comes through tomorrow in terms of the backlog reported there which, again, is rounded. It's shown in gigawatts and billions. So there's a lot of error that can happen in that rounding. But that's where we've actually been booking new bookings. So there's a couple of atypical events that happened, as we mentioned on the call, that impacted the number you'll see in the Q. And that was this customer dispute accommodation we had and the discrete events around Cove Mountain and Muscle Shoals. But relative to where we're seeing the market now for new bookings, so it's up, we're seeing bookings out with a freehand, and we're very happy with where we're seeing those.

--------------------------------------------------------------------------------

Mark R. Widmar, First Solar, Inc. - CEO & Director [10]

--------------------------------------------------------------------------------

Yes. And just -- and you got to remember, too. Just the natural cadence is going to be that, as we ship, we're shipping '19 volumes and we're replacing them with '21 or later volumes. You're going to see some natural erosion of that metric. But at the same time, you'll see margin expansion because the cost profile that we'll have in that horizon is going to be much more advantaged relative to what we're -- what our current cost of our Series 6 production is. So you got to look at it from both perspectives. But the metric in and of itself naturally is going to trend down over time.

--------------------------------------------------------------------------------

Operator [11]

--------------------------------------------------------------------------------

Your next question comes from Brian Lee with Goldman Sachs.

--------------------------------------------------------------------------------

Brian K. Lee, Goldman Sachs Group Inc., Research Division - VP & Senior Clean Energy Analyst [12]

--------------------------------------------------------------------------------

I'm going to try to squeeze in 3 here as well. I guess first off, in the past, you've been telegraphing some bookings moderation moving to the year. We're obviously not seeing it yet. So maybe just a quick update on your thinking around bookings trajectory as you move through 2019. Has that changed versus what you said before? Second, on the gross margins, you mentioned Cove Mountain and Shoals a bunch, Alex, and I know they're moving around in terms of buckets. But was that in the modules gross margin of 5%, or was it embedded in the systems gross margin? And then can you quantify what that impact was?

And then just last one I'll squeeze in here. On the 5% to maybe 7% nameplate capacity increase at Ohio, Mark, can you elaborate a little bit more? Is that a debottlenecking process? What amount of CapEx is involved? And then is it a quarter or 2 if you see success where you could then decide to roll it out to Malaysia and Vietnam?

--------------------------------------------------------------------------------

Mark R. Widmar, First Solar, Inc. - CEO & Director [13]

--------------------------------------------------------------------------------

Yes. I'll do the bookings and the capacity and let Alex talk to the gross margin. But I've been real happy with the ability of our sales team to engage with customers and to get to an opportunity to book volumes that are sitting that far out in the horizon. But one of the things, as I said in the prepared remarks, is that there is a significant undertone of wanting to lock in to multiyear agreements, having certainty around the technology and ability to deliver the committed bin at which they're going to rely on and then design around and the fact that we'll stand behind the contracts, and there's not going to be repricing or anything else along those lines. And I think that's playing to our strength. The 2 large orders that we have for the first half of this year: one in Q1 and the other one here. And even the one that we said that there's a Phase 1 and a Phase 2 really is reflective of that. And it's really the confidence in developing those types of relationships with our partners and, obviously, a shared commitment and trust. And having gotten in certain situations -- been put in difficult situations by some of our competitors at various points in time. And given safe harbor and other things that are in front of us, nobody wants to be in that type of position where they have a partner that is unable to supply them or deliver against their commitments. So I think that's playing to our strengths. So as we have sort of given an indication of second half-weighted bookings, we clearly are ahead of where we thought we would be. There's still a lot of opportunity even yet for the balance of this year. As we said, we still are confident in our ability to get to at least the 1:1 booking ratio, which I would say we've got another 1.2 gigawatts or so, something in that range. We've got a couple of pretty significant deals that hopefully we can be able to bring across the finish line over the next few quarters. So I think we can get there. But yes, I've been very pleased with what we're seeing from that standpoint.

On the capacity side, it's really -- it's a debottlenecking. It's also an -- it's basically -- there is some CapEx but you think about where your constraint is in your production process and then adding a little bit of capacity there or resiliency, redundancy, and that creates incremental throughput across the remaining operations. And so there's a little bit of that. It's not a significant CapEx. And if we are successful, we can roll it out within a matter of a few quarters across the balance of the fleet. The only -- the real lead time issue is just to the extent there's capital requirement, it's just the lead time from the vendor to provide the capital, but it's a relatively short time from the point of decision to actually seeing the benefit of the throughput.

--------------------------------------------------------------------------------

Alexander R. Bradley, First Solar, Inc. - CFO [14]

--------------------------------------------------------------------------------

Brian, so the question on gross margin. If you look at how we typically structure deals historically when we had an EPC agreement there's a profit pool associated with development, a profit pool associated with the module and then a piece with the EPC. Those are all lumped together into one large profit pool. It would then be recognized on a percentage of completion basis throughout the project life. What we've done here is we've sold the development project asset and then fully recoup our development cost and take our development fee upfront. And remember, if you look at the risk profile that we talked about at our Analyst Day 1.5 years ago, that's a higher margin piece of the business. And that's flowing through the systems segment. And then now we then have a module agreement with that project company to deliver modules, and that module revenue and margin will flow through the module segment over time. And so that's the split you're seeing here. We haven't broken out the exact delta, but that's how it's going to be recognized over time.

--------------------------------------------------------------------------------

Operator [15]

--------------------------------------------------------------------------------

Your next question comes from Michael Weinstein with Crédit Suisse.

--------------------------------------------------------------------------------

Michael Weinstein, Crédit Suisse AG, Research Division - United States Utilities Analyst [16]

--------------------------------------------------------------------------------

Can you -- given your visibility in the Safe Harbor demand and pricing, do you expect to extend the Series 4 lines through the first quarter of '20?

--------------------------------------------------------------------------------

Mark R. Widmar, First Solar, Inc. - CEO & Director [17]

--------------------------------------------------------------------------------

We're looking at Series 4 and opportunities for that. We have gotten requests from customers around availability of Series 4. You got to remember, for sure, right now we're running 2 factories in Malaysia -- excuse me, of Series 4. We refer to them as KLM 1 2 and KLM 3 4. 3 4 for sure is close -- is shutting down. We're going to ramp that up. We're going to get Series 6 production out of that product most likely by the end of 2020. It's a matter what do we do with KLM 1 2. And there's a couple of different things. There's some of them here domestically in the U.S. and we've got some pull from customers right now to do a few things. Safe harbor as well as -- we have some of the examples, which customers are actually having issues with damaged modules, hail damage, in particular and asking if there's a way, not for solar technology, but crystalline silicon technology, and asking if there's a way that we could probably help support their needs where they may actually replace some of the crystalline silicon and use a First Solar product. We only have Series 4 that can actually accommodate that type of request. There's other opportunities with pull-through of opportunities internationally for Series 4 in some key markets, so we're looking at that. And there's some synergies that we capture. Think of it as even purchasing power across the glass while the actual form factor of the glass is not the same between Series 4 and Series 6. There are some synergies along that. Aggregate purchasing power, volume breaks and the like that we could maybe get some additional leverage not only for better pricing on our cost on Series 4 but potentially better cost on Series 6. So there's a lot of options that are being looked at. We'll have more information around that probably in the upcoming guidance call that we'll do by the end of the year.

--------------------------------------------------------------------------------

Operator [18]

--------------------------------------------------------------------------------

Your next question comes from Colin Rusch with Oppenheimer.

--------------------------------------------------------------------------------

Colin William Rusch, Oppenheimer & Co. Inc., Research Division - MD and Senior Analyst [19]

--------------------------------------------------------------------------------

So last quarter, you talked a little bit about the project business and some inefficiencies there that you were working on fixing. Can you just give us an update on what's happening there and what you're able to accomplish in the last quarter?

--------------------------------------------------------------------------------

Mark R. Widmar, First Solar, Inc. - CEO & Director [20]

--------------------------------------------------------------------------------

Yes. So there's -- I think what we highlighted and I think it was in the last earnings call or maybe the one before that, we have put a new leadership team in place. And we are evaluating the kind of long-term relative competitive position of our own EPC offering and trying to determine what is the right strategy. I mean the view is very simple. If we can be best-in-class as it relates to our EPC execution, then similar to what we feel like we're best-in-class in O&M, we're best-in-class on our modules technology, if we can do the same with EPC, then that will be the path that we'll pursue. If we can't, then we'll have to look at options. As Alex indicated in his remarks, we have structured 2 deals here recently that we're only in the sale of development asset with a follow-on model sale agreement. We chose not to engage with EPC execution at this point in time, just given the uncertainty and the changes that we're going through until we can get it in our feet back underneath us. It's hard to sort of make additional commitments. As it relates to the existing projects and the execution against the existing projects, I think the team has made some very good progress getting where we need to be on cost and schedule. We still need to go -- we still need further to go in terms of our ability to meet and satisfy our customers' expectations, whether it's third-party EPC. For example, when we -- I would say that we haven't necessarily lived up to a standard that we want to live up to in terms of customer experience. And we got to do a better job from that standpoint. But look, there's a lot of moving pieces that we're looking at in evaluating a number of different options, and we'll love to give you more color when we have more details.

--------------------------------------------------------------------------------

Colin William Rusch, Oppenheimer & Co. Inc., Research Division - MD and Senior Analyst [21]

--------------------------------------------------------------------------------

Great. And then just one quick follow up. Just if you look at the mid- to late-stage pipeline, how many new customers are in that? And how much of that is really existing customers that may try to leverage some of that potential new business into price reductions on existing contracts?

--------------------------------------------------------------------------------

Mark R. Widmar, First Solar, Inc. - CEO & Director [22]

--------------------------------------------------------------------------------

I don't have the -- look, the 2 orders that we got for this quarter were new customers. I don't have that -- the exact mix between what sits in there in terms of existing customers versus new customers. I would argue though that there is a pretty high percentage of that pipeline that are customers that we've done business with. That's largely representative of what you see in the marketplace. It's the NextEras of the world, the EDFs of the world and others. I mean they're going to be key customers. And they'll be not only in our contracted volume but they'll be in our pipeline volume as well.

--------------------------------------------------------------------------------

Operator [23]

--------------------------------------------------------------------------------

Our final question will come from the line of Travis Miller with MorningStar.

--------------------------------------------------------------------------------

Travis Miller, Morningstar Inc., Research Division - Director of Utilities Research and Strategist [24]

--------------------------------------------------------------------------------

You answered most of my questions, but I had a higher-level question. You mentioned batteries in your remarks there. And I'm just wondering how does that impact you guys? Is battery development is something that improves the number of megawatts you can sell? Or does it change pricing at all? Is it something that you might even eventually be involved in, in terms of offering full package?

--------------------------------------------------------------------------------

Mark R. Widmar, First Solar, Inc. - CEO & Director [25]

--------------------------------------------------------------------------------

Yes. So I mean, it's obviously clear. I mean as it relates to the battery, it clearly enhances the overall solar value proposition. We've talked about this before, as we've moved from energy-only contracts to, we refer to it as flexible solar generation, which allows you to effectively provide value beyond the energy, which could include ancillary services and we've been able to leverage that in some key markets and some opportunities. And we've done a number of studies, whether it -- we've done a study with Cal ISO and NREL. We've done a study with E3 and TECO around that, demonstrates kind of that flexible solar value proposition. And then it gets into dispatchable energy. And so that's the evolution that we see. And we do believe that you can get to a relatively high solar penetration in a number of key markets before you ultimately have to get to battery integration. But the market's starting to trend that way. We do think it sort of creates this disruptive opportunity where you can displace the deal that we won with APS and what we refer to as our Sun Streams 3 project. That was on all resource. We competed solar against gas peaker and other forms of generation, and we were able to win a portion of that RFP. So I think it just further enhances the overall value proposition.

Now as it relates to technology, I mean the technology as it evolves right now is dominated by lithium ion. And it's really leveraged off of the scale. It's been created through EVs. And I see that as kind of the near term the most competitive solution that will be in the marketplace. So I wouldn't expect us to get into the battery side of it. Now on the power plant control side and the optimization and dispatching of the energy generation and optimizing that against the -- for example, what we've done with APS, there's a pricing tier structure that determines the energy value that's being generated and how do you optimize charging of the battery versus dispatching of the battery to optimize and to capture the highest value. Those are the things that we'd like to stay close to. We think that fits in well. We'll do O&M on that plant. We already have the power plant control, so it's more or less modification of the capability we already have that further enhances the value proposition that we can provide to the customer. So that's how we see. It's an important piece, it will evolve. Near term, I think you can see a quite high solar penetration before you have to get into dispatchable generation. But in some markets like California, the value proposition is a little bit different and more immediate need potentially in points of time throughout the year, that we'd see more battery installations there. You'll see some in Nevada, you'll see some here in Arizona. But a number of other markets are probably way too early in their evolution to really require it at this point in time.

--------------------------------------------------------------------------------

Operator [26]

--------------------------------------------------------------------------------

There are no further questions at this time. This concludes today's conference call. You may now disconnect.