First Solar Inc

NASDAQ:FSLR

Earnings Call Transcript 2019-Q1

[Operator]

Good afternoon, everyone and welcome to First Solar’s 2019 Guidance Call. This call is being webcast live on the Investors section of First Solar’s website at investors.firstsolar.com. At this time, all participants are in a listen-only mode. 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.

[Steve Haymore, Investor Relations]

Thank you. Good afternoon, everyone and thank you for joining us today. The company issued a press release announcing its guidance for 2019. 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 the 2019 outlook. Following their remarks, we will then have time for questions.

Please note that 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?

[Mark Widmar, Chief Executive Officer]

Thanks, Steve. Good afternoon and thank you for joining us today. Before discussing recent business update and our outlook for 2019, I want to provide a perspective on the progress that we have made on our Series 6 transition since our Analyst Day event last December. At that time, we displayed our first complete module, which was a combination of more than a year’s worth of effort to develop, test and solve and qualify hundreds of neutrals. While completing our first Series 6 module as an important achievement, what we have been able to accomplish in 2018 is even more noteworthy. Over the course of this year, we have started up three Series 6 factories. And year-to-date, we have produced over 1.4 million modules.

In addition, our fourth Series 6 factory is scheduled to start production by the middle of Q1 2019 and at this facility, which is under construction, is planning to start production at the beginning of 2020. To concurrently manage all the activities related to the construction, startup and ramp of five different factories, there is a major undertaking and to reach the point we are at today is a significant accomplishment. As we closed the year, we are continuing to push forward on tool upgrades and other process improvements that will position us to meet our throughput and efficiency goals in 2019. In addition to these manufacturing achievements, we have also had a notable success booking new systems and module business this year. As we mentioned during our earnings call in late October, our year-to-date net bookings were 5.2 gigawatts, which included over 1.6 gigawatts of systems business.

I would not be providing a complete update on our bookings activity during the call today. I want to highlight some of our recent success, especially as it relates to our project development business. In the past 2 weeks, we have signed 2 new PPAs, which combined are more than 300 megawatts DC. The first PPA signed was 120 megawatt AC expansion of our Little Bear project with MCE, a leading community choice aggregator in California. In late 2016, we announced the initial 40-megawatt AC PPA with MCE and now the necessary load requirements, have been met that our project will expand into 160 megawatts AC.

Community choice aggregators, or CCAs, are increasingly important providers of electricity to consumers both in California and in a growing number of other states. CCA has offered their customers the benefits of local control, which is driving an increased deployment of renewable energy. We are excited to continue our association with MCE and support their deployment of cost effective solar energy. We also recently signed a greater than 120 megawatt AC PPA with a major utility customer in the Western United States. This project will support collaboration between the utility and its corporate buyers to meet their renewable energy objectives. More details related to this project will be available at a later date and we expect this project will come on booking following the resolution of one outstanding condition presented.

This most recent signed PPA highlights the strong demand that we are continuing to see from corporate customers for utility scaled solar. In the past 3 months, we have signed more than 500-megawatt DC of PPAs with utilities or corporate customers are the intended customers of the energy to be generated by these projects. This brings our total contracted systems projects with corporate off-takers to nearly 700 megawatts DC. As I indicated in our last earnings call, we believe demand for solar power like corporate customers will continue to grow in coming years based on the favorable economics of solar power, as well as the wind production tax credit continues to step down.

Turning to Slide 4, I’ll discuss in greater detail our growing portfolio of contracted systems projects, which includes the new PPAs announced on today’s call. In reviewing the slides, keep in mind, that it is based on the timing of model shipments, which does necessarily correspond to the specific year in which revenue is recognized. As highlighted on the slide, contracted shipments after 2018 are more than 2.4 gigawatts DC. Adding in mid-to-late stage opportunities of 1.6 gigawatts DC brings the total potential future shipments to 4 gigawatts DC.

Keep in mind, this does not include our larger portfolio of early stage opportunities. An important point to note is the contracted systems shipments in both 2019 and 2020, which are approximately 1.2 gigawatts and 1.1 gigawatts respectively, already exceed our target 1.0 gigawatts per year of systems business. With over 300 megawatts of prospective bookings opportunities in 2020, we have the potential to exceed our annual system shipment target of 1 gigawatt by an even greater margin. In 2021, we already have nearly 1.3 gigawatts of mid-to-late stage development opportunities and over time, we’re anticipating adding more of our early stage opportunities to this number.

Now as we’ve emphasized in the past, we continue to have a strong portfolio of systems projects in Japan. This includes over 160 megawatts DC of book projects that will be solid in 2019 or later and a mid-to-late stage pipeline of projects totaling 280 megawatts DC, which are not yet considered bookings. Of 280 megawatts of this pipeline have fixed feed-in-tariff arrangements, but the projects are not yet included in our contracted volume and in the completion of certain development activities, including interconnection agreements.

As it relates to US PPA opportunities, we plan to leverage our balance sheet capacity in 2019 to meet the ITC Safe Harbor requirements for a substantial portion of our development portfolio, which allows us to preserve the 30% ITC through the end of 2023 for these projects. Alex will provide more detail on the financial requirements for this strategy, but we expect to be able to preserve value of up to 5 gigawatts DC of contracted and uncontracted projects. We anticipate that taking advantage of this opportunity will not only add value to the projects in our contracted portfolio, but will enable more competitive bidding and higher project values for uncontracted side.

Continuing on to Slide 5, I’ll discuss of capacity plan for 2019 and 2020. As communicated previously, our current roadmap outlines plan for 6 Series 6 factories with the possibility of additional expansion depending on future market conditions. With our second Vietnam factories scheduled to start production in the middle of next quarter and the second U.S. factories scheduled to begin operation at the beginning of 2020, we anticipate having 5.4 gigawatts of Series 6 nameplate capacity by early 2020. In order to best position the company for entering 2021, our current roadmap plans for the shutdown of our 2 remaining Series 4 factories in Malaysia at the end of 2019.

Now to enable this roadmap, we have utilized the flexibility in our existing customer contracts to transition them from Series 4 volume in 2020 to Series 6. Once shutdown, we expect to immediately begin converting one of these factories to a Series 6 facility with the nameplate capacity of 1.2 gigawatts and the start of production target for the end of 2020. This would bring our Series 6 nameplate capacity to 6.6 gigawatts entering 2021 and also provides the option to grow nameplate capacity to 7.8 gigawatts in 2021 or later if we convert the second Series 4 factory. The decision of whether to convert this last Series 4 factory to Series 6 will depend on market conditions and visibility to demand. Note that in addition to the ability to convert our final Series 4 factory in Malaysia to Series 6, we also have an option to add Greenfield factories at our other manufacturing locations in Vietnam and Ohio.

Next, Slide 6 provides production level in 2019 and 2020 based on the capacity roadmap discussed. Total production in 2019 is planning to range between 5.2 and 5.5 gigawatts, which includes approximately 2 gigawatts of Series 4 production. In 2020, total anticipated production is similar to 2019 at 5 to 5.5 gigawatts, but at this point, all volume will be Series 6, our most advantaged product. The ranges of potential Series 6 production we are showing in 2019 and 2020 are a reflection of the variation that can occur in factory startup and ramp timing. To the extent we are able to start production earlier than scheduled or increase throughout levels that ramping factories faster than anticipated, we should expect to reach the high end of these ranges. As a reminder, Series 6 production is expected to increase significantly over this time period from approximately 700 megawatts in 2018 up to 3.5 gigawatts in 2019 and then up to 5.5 gigawatts in 2020. This leaves to a total combined 2019 and 2020 production of between 10 and 11 gigawatts. At the midpoint of this range, approximately 90% of this production is already contracted and we have good line of sight to booking the remaining available supply, most of which is in 2020.

In summary, we are encouraged by our position as we end our 2019 with strong demand visibility and a capacity roadmap that positions us with growing levels of Series 6 capacity. I will now turn the call over to Alex who will provide more detail on our 2019 outlook.

[Alex Bradley, Chief Financial Officer]

Thanks Mark. To start, I would like to add to what Mark discussed at the beginning of today’s call regarding the progress we have made on our Series 6 transition. From a balance sheet perspective in late 2016 when we made the decision to accelerate our Series 6 plans, we had a net cash position of $1.5 billion. We have since invested approximately $1 billion in Series 6 CapEx and today have a net cash position that has grown to over $2 billion. The ability to not only maintain, but increase our liquidity, highlights the strong execution that’s taking place throughout the transition.

Before covering 2019, I will briefly discuss our 2018 guidance. As we indicated in our call in October, we anticipate selling two projects in Japan in Q4. Although they have not yet closed, we continue to expect to close both deals before the end of the year. However, if the timing of these sales were to push for 2019, we will fall below the previous earnings guidance provided for 2018 and there will be an offsetting benefit for 2019 guidance provided on today’s call.

Turning to Slide 8, I will discuss our guidance assumptions for the upcoming year. Starting with production, our output next year is expected to increase by more than 80% compared to 2018 as a result of planned increase in Series 6 volume. As mentioned earlier, 2019 production is targeted to be between 5.2 and 5.5 gigawatts. That will consist of approximately 2 gigawatts of Series 4 production between 3.2 to 3.5 gigawatts of Series 6. 2019 volumes sold is expected to be more than 5 gigawatts, including approximately 4 gigawatts of module sales and more than 1 gigawatt of system sales.

Sales of development projects are planned to comprise the majority of the system sales. The mix of 2019 net sales is anticipated to be approximately 40% to 45% module and 55% to 60% systems. The mix of module versus systems revenue provided is consistent with our segment reporting view where any modules used in projects are reflected in the systems segment. Our ongoing Series 6 capacity expansion is expected to impact 2019 operating income by $110 million to $113 million. This is comprised of $20 million to $30 million of ramp costs, which is through existing cost of sales and $90 million to $100 million of plant startup which is treated as an operating expense. Manufacturing ramp costs are projected to be incurred primarily in the first half of 2019 and are driven by our two factories in Vietnam. The first factory, which started production in September, will have residual ramp costs into the first part of 2019 and the second Vietnam factory is anticipated to transition from startup to ramp in Q1. Plant startup expense is projected to be weighted towards the second half of 2019.

Firstly, we expect startup expense at our first Vietnam factory prior to the aforementioned transition to ramp in Q1. Secondly, we expect significant startup expense in Perrysburg as we bring online our second Series 6 factory over the course of 2019. The Ohio factory startup is estimated to be our most expensive to-date due to the geographic location and the transition of our existing Series 4 workforce factory. While the transfer of the workforce will take place sooner than otherwise will be optimal for the Series 6 plant operations, it allows us to maintain a highly skilled workforce and in 2020 we should reap benefits of this decision with a fall of the manufacturing ramp. Thirdly and the reason the plant startup is higher in 2019 than in 2018, we are investing to bring in-house certain operations that are anticipated to reduce the Series 6 module bill of materials. While it’s increased its startup expense in 2019, resulting benefit from these initiatives will be realized in 2020 and beyond in the lower module cost per watt.

Final valuations of these programs will be completed in coming months and startup expense 2019 could be up to $15 million lower if we determine there is not a sufficient return associated with these projects. While we are not providing specific guidance around the Series 6 module cost per watt for 2019, we do anticipate significant improvement over the course of the year. The planned reduction in cost per watt is projected to result from better throughput, further reductions to the bill of materials and efficiency improvements. Keep in mind that the most significant reductions in Series 6 cost per watt should come in the second half of the year when factory utilization rates will be at the highest.

Lastly on this slide, we planned to meet the ICC Safe Harbor requirements in 2019 in order to preserve the 30% ITC of up to 5 gigawatts to development project sites, a portion of which will include PD Plus storage. Some of these sites already have contracted off-take agreements while others are not contracted and will be bid into future PPA opportunities. The 2019 investment is targeted to be between $325 million and $375 million as effectively an acceleration of project inventory who would otherwise have been procured in 2020 or later. While this pull forward in spending will lower our operating cash flow and net cash balance in 2019, we anticipate a significant return on this investment in the form of higher project values with Safe Harbor inventories utilized.

On Slide 9, we provide a review of key projects, comprised most of the approximately 900 megawatts DC of development sales we expect to recognize in 2019, included in this list of projects to be sold next year such as Sun Streams, Sunshine Valley and Seabrook in the U.S. and internationally in Chicago and Japan and various India assets. We also plan to recognize most or all of the remaining revenue on certain projects that were sold in 2018, including the GA Solar 4 and Rosemont projects in the U.S. and Beryl internationally. While not included on this list because it’s an EPC agreement, we also anticipate recognizing most of the revenue from the 250 megawatt AC Phoebe Texas project next year. Keep in mind that as in the past, the timing of projects sold out is subject to a variety of factors, which can ultimately impact on the recognized revenue. Beyond 2019, as Mark noted, we have a strong contracted development pipeline and a portfolio of opportunities that we are actively pursuing in order to our systems bookings in 2020 and 2021.

On Slide 10 we have an updated view of our operating expense profile excluding production startup. This trend highlights our effectiveness in scaling our largely fixed OpEx structure as we plan to increase production levels in 2019. Total operating expenses are increasing relative to 2018 due to a higher investment in R&D, sales and other resources necessary to support our growth plans. The increase in R&D is to fund certain specific advanced programs which will be evaluated on an ongoing basis and did not necessarily reflect the long-term spending rate. While total OpEx dollars are increasing on a per watt basis, 2019 OpEx is expected to decrease to $0.06 per watt, a reduction of one-third versus 2018. The significant improvement in our OpEx per watt highlights the benefits of scaling we have discussed frequently over the past 2 years and ultimately results in improved operating income. Beyond 2020 as we ramp to our full nameplate capacity of 6.6 gigawatts and potentially add more capacity beyond that, we would expect OpEx per watt to continue to reduce.

I will now cover the 2019 guidance ranges on Slide 11. Our net sales guidance is between $3.25 billion and $3.45 billion with sales expected to be slightly more weighted to the second half of the year. Gross margin is projected to be between 20% and 21% with the module segment margin higher than the systems segment margin. Gross profit is anticipated to be significantly more weighted to the second half of 2019 for several reasons. Firstly, ramp cost of $20 million to $30 million which are included in cost of sales are planned primarily for the first half of the year. Secondly, module cost per watt excluding the impact of ramp costs is anticipated to improve over the course of the year as throughout and efficiency increase.

To illustrate this point, by Q4 of 2019, we expect our fully ramp Series 6 factories to have a module cost per watt that is 40% lower than our 2016 Series 4 cost per watt adjusted for a $0.01 per watt cost impact related to the higher cost of the frame. This is essentially in line with the original cost reduction target we set over 2 years ago at the outset of this transition. Thirdly, we expect our higher sales from EPC projects which carry lower margins than our development projects in the first two quarters of 2019. This means that we will have higher development project sales in the second half of the year.

2019 operating expenses are targeted to be between $390 million and $410 million, which includes production startup expense of between $90 million and $100 million. We anticipate R&D expense of approximately $105 million and SG&A expense of $200 million in the midpoint of our guidance range. Operating income is estimated to be $260 million and $310 million as the increase of between $110 million and $130 million of combined ramp costs and plant startup expense.

One item to highlight for 2019 is the uncertainty surrounding our class action lawsuit that was filed in 2012, including the possibility of higher legal costs related to our ongoing defense for the class action. Until this point, essentially all of the cost to defending the case has been covered under insurance policies. But given the number of years of litigation has been ongoing and with the potential with the case that goes trial in 2019 we may well exceed our insurance coverage limits. As it relates to the case in October, the U.S. Supreme Court referred them as to the U.S. Solicitor General in connection with our pending petition for CERC. We are awaiting the recommendation of the Solicitor General. The timing of recommendation is uncertain as is any final decision of the Supreme Court whether to accept or review the case. Our guidance figures do not take into account the financial impact of the continued litigation and any resolution of the lawsuits.

Turning to non-operating items, we expect net interest income of approximately $40 million and no contribution from equity and earnings. Factory expense is estimated to be approximately $50 million. The resulting earnings per share of $2.25 to $2.75 is projected to be weighted towards the second half of 2019 due to the factors impacting gross margin already discussed.

Turning to cash, operating cash flow which is projected to be approximately breakeven in 2019 has impacted significantly by two items. Firstly, when we sell an asset with project level debt that is assumed by the buyer, the operating cash flow associated with the sale is less than its supply had not assumed the debt. In 2019, we expect buyers of project to assume approximately $350 million of liabilities relating to these transactions. The second item reducing 2019 operating cash flow is $325 million to $375 million that we plan to invest in Safe Harbor inventory. Capital expenditures in 2019 are projected to range from $650 million to $750 million, including approximately $600 million for Series 6 capacity expansion. Majority of the CapEx is associated with our second Vietnam and Ohio Series 6 factories. The remainder of the CapEx is dedicated to investment in R&D and initiatives to improve module cost per watt and enable the scaling of our business. Note that for the full year, we expect depreciation expense of approximately $200 million. Ending 2019, net cash balance is anticipated to be between $1.6 billion and $1.8 billion. The decrease of our estimated ending 2018 net cash balance is primarily due to the investment in Safe Harbor inventory and capital expenditures partially offset by cash flows associated module and project sales.

As it relates to how we view liquidity in deploying our balance sheet in 2019, we continue to see value and flexibility from the strong cash position. Some examples of how we plan to use this flexibility to take advantage of opportunities in the global market that’s expanding are as follows. Firstly, in 2019, we plan not only to continue investing in our announced Series 6 capacity, but a strong balance sheet provides us with the option to commit to investing in additional manufacturing if market conditions warrant.

Secondly, it provides us with the ability to invest in and grow our systems business. A few months ago prior to the ITC commenced construction guidance, we would not have anticipated the need to invest in Safe Harbor inventory in 2019, but by maintaining balance sheet flexibility, we were able to pursue this opportunity which has the potential to generate attractive return. Another reason to maintain this flexibility is that similar to 2018, there maybe opportunities next year to acquire development projects or pipelines. Returning capital shareholder is something we will continue to evaluate during the course of 2019. This evaluation will take into account the need to maintain a minimum liquidity position, the need to maintain the flexibility for new investment opportunities and also consider the potential resolution of the class action lawsuit.

Turning to Slide 12, we have provided a summary of all expenditures related to Series 6 capacity expansion over the next 2 years. As previously indicated, 2019 combined ramp cost and plant startup expenses are $120 million at the midpoint of the range. In 2020 based on our current roadmap of 6.6 gigawatts of Series 6 capacity, we expect to combine total of startup and ramp to decline significantly. Startup is planned to decrease to $30 million and associated with our second Series 6 factory in Malaysia which is anticipated to start production at the end of 2020. Ramp costs are estimated to be $25 million in 2020 primarily driven by our second Ohio factory, which starts production at the beginning of 2020. So between $600 million and $650 million of Series 6 CapEx planned for 2019, cumulative Series 6 CapEx would reach between $1.7 billion and $1.8 billion at the end of next year. The total targeted Series 6 spend is $1.9 billion for the initial 6.6 gigawatts of capacity that’s been committed, which leaves between $100 million and 150 million of CapEx in 2020. And keep in mind that these numbers incorporate CapEx only for Series 6 and do not include capital of R&D or other corporate purposes. In addition to the extent that we commit to expansion beyond the initial 6.6 gigawatts of capacity, CapEx started on ramp numbers in 2020 would increase.

Turning to Slide 13, I will summarize the key messages from today’s call. Firstly, we continue to make significant progress on our Series 6 transition from both supply and demand perspective. On the demand side, our systems pipeline is strong. We have contracted bookings to underpin nearly all of our production over the next 2 years. In 2019, we expect strong system sales of over 1 gigawatt, including approximately 900 megawatts of development project sales. Module sales are expected to grow significantly in 2019 to approximately 4 gigawatts. With this growth in production, sales will continue to scale our largely fixed OpEx profile and expect to lower OpEx forward to $0.06 in 2019.

Secondly, we expect solid revenue growth in 2019 based on our guidance of between $3.25 billion and $3.45 billion. Our earnings per share guidance is between $2.25 and $2.75 in 2019. And our earnings are impacted by combined startup and ramp costs of $120 million this investment enables expansion of our Series 6 capacity positioned us for future growth. And thirdly, our balance sheet continues to be stronger in transition and provides us with the ability to continue to invest in Series 6 manufacturing capacity and with significant flexibility to take advantage of other opportunities across the value chain.

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

[Operator]

Thank you. [Operator Instructions] And the first question will come from Philip Shen with ROTH Capital Partners.

[Philip Shen, ROTH Capital Partners]

Hi, good morning. Thanks for the questions. The first set of questions is around the margin guidance. It appears a little bit light, I know you are improving your throughputs and it sounds like you are going to get to your ideal cost structure by Q4. So, one, want to confirm that and then two I want to see if there is a chance that whether or not that could happen earlier than Q4 and is the margin kind of outlook – is the challenge there as a result of the throughput issues is a little bit lower than expected? In terms of bookings, I know you guys have talked about 2 gigawatts out there for 2019 and you used to have some for 2020, is the idea for 2019 – is there any risk in the 2019 Series 4 bookings could they be shifted to Series 6 or our people pretty locked in there and as it relates to 2021, can you give us some visibility into how those bookings are coming? So I appreciate addressing all these questions and thanks very much.

[Alex Bradley, Chief Financial Officer]

I will start on the margin piece and Mark can expand on market expansion and talk about the Series 4, but on the margin side, I think from a dollar basis, we’re very happy with where we are. The gross margin guidance of 20% to 21% is a blend obviously of the module business and the systems business. The systems gross margin is a blend of EPC and self-developed business. Self-developed asset – systems business is still in the range of 15% to 20% on a total basis, which is consistent with what we said historically.

If you go back to our last Analyst Day, we try to break it out into constituent parts and when you add it those all up in aggregate, you’ll get to that 15% to 20% range on the systems side. So you’re seeing slightly lower margins on the self-developed systems side than you are on a blended basis, which implies the module margins obviously are higher in that blended 20% to 21%. You’re also seeing a significant amount of EPC business in there, which is again lower margins on a percent basis, still accretive on a dollar basis and still additive to the year, but it’s always going to be dilutive on a percent basis when you look at it on a total percentage basis. So I’d say look we’re pleased with where we are in terms of margin on the Series 6 module business and the system business is coming in, in line with what we thought it would be, but what you’re seeing I think is a little bit higher revenue coming from some of the EPC business, which again is dilutive on a percentage basis, but still accretive on a total gross profit dollar basis.

[Mark Widmar, Chief Executive Officer]

Yes, it is – around that I would just say, Philip, if you look at the original business case that we put together for Series 6 gross margin percent, we’re actually going to be on an average and even for the – and particularly, the energy rate will achieve or exceed our expectations relative to the original business case. So Series 6 combination of ASP and module cost is exceeding our expectation. Again, if you look at for the average for the year and particularly if you look at it for the exit rate, the margin percent is much higher than we originally anticipated. So we feel very comfortable and very good about having that position on Series 6 and leveraging as we go into 2020. Bookings around, so I think your question was moving customers from 4 into 6 or any risk of that happening. Just – I want to make sure it’s clear. One of the comments that we’ve made in the prepared remarks is that, we have already shifted our Series 6 volume in 2020, Series 4 volume excuse me and contracted in 2020 to Series 6. So we’ve already worked with our customers. We have flexibility in contracts. We’ve referenced that before that for particular customers we could move them into Series 6. So we already have made that shift. And again, part of the reason we want to do that, it allowed us to ramp down some Series 4 production faster and then allows us to create greater optionality for more volume of Series 6 in 2021 in particular.

So we completed that. As it relates to Series 4 going into Series 6 for 2019 shipment, where we are right now, that’s all contracted committed, customers have already designed their projects for that product. It’d be very difficult for us to make any changes to our customers at this point in time because of redesign and other efforts that would be necessary for our customers. It also means we’re constrained to make that shift, but they’re also constrained to try to look at other options relative to the current Series 4 product that they are contracted, they’re obligated to take in 2019. So I don’t see any real movement there. Again, we have addressed it for 2020, but really probably not a lot opportunity to do anything in 2019. 2021 bookings, I continue to be very encouraged and I’ve been traveling a lot, I’ve been in Europe recently, in Asia, and what I continue to hear from customers is very strong demand, willingness to talk into multi-year horizons that do go out into 2021. I would expect us, we booked some 2021 volume in our last quarter, last quarter call north of gigawatt of bookings. We did start to put some points on the board for 2021. I think there is a reasonable opportunity that we’ll be able to close on a couple of large orders that we’re negotiating right now with customers that would start to fill up that ‘21 bucket and that’s really what we’re focused on right now. As we indicated, we’re essentially sold out all our volume through 2020 and the negotiation, the conversations we’re having with our customers right now is how do we start selling through in ‘21 and beyond.

[Operator]

And our next question will come from Brian Lee with Goldman Sachs.

[Brian Lee, Goldman Sachs]

Hey guys, thanks for taking the questions. Maybe just to follow-up real quickly on that first question from Philip. I may have missed it, but if I look at Slide 4, most of your systems volume for 2019 looks like it’s in that development bar not EPC. So just wondering, it’s – the margin comments sounded like the EPC mix was what was weighing on the ‘19 margins for systems, but if you could just help elaborate a bit there?

[Alex Bradley, Chief Financial Officer]

Yes. So if you look at that bar, you’re seeing about 900 of development business and about 300 roughly of EPC business. So that the self-developed business is coming in around the margin guidance we expected. And if you look back at the numbers we gave at the Analyst Day last year, those roughly hold and if you blend those together on a indicative capital structure for projects, you’re going to get somewhere in that 15% to 20% overall gross margin range on the self-developed assets, but that 300 megawatts of EPC business although the module in there is still good margins, the rest of that EPC business is still in that 5% gross margin range. So that does weigh down across on a blended basis.

[Mark Widmar, Chief Executive Officer]

Yes. But maybe I add a little bit more to that, Brian is, is that the total gross margin 20%, north of 20%, alright, so what we’ve said before is that our systems business would be in the range of 15% to 20%, and self-developed would be at the upper end of that range at 20%, and EPC would be at the lower end, 15%. Now again, that includes the module, right, so put that – just make sure we put that in perspective, right. So you got the module in both for systems business and both the EPC business, high-end 20% for systems, self-developed, and low-end 15% for EPC.

The other thing that we said is that we’re still selling through Series 4 and we are happy with the margin realization on Series 4. But again, we pointed to that to be in kind of the mid-teen – mid-to-upper teen, so that’s going to also say that, that number is below the 20% and that’s 2 gigawatts of our volume. So when you look at the development business, the EPC business, and the Series 4 being below the average, you would – it clearly points to very strong gross margin on the balance of that volume being Series 6 third-party module sales in order to blend upward from what that average would represent on a total revenue basis. So again, so hopefully, that helps a little bit without giving the specifics, but we’ve indicated previously around margin percent on Series 6, we’re at or above that number. And as we trend towards the year, as the year starts to progress, we’ll see even better margin realization around Series 6 and all those expectations are at or above what we’d originally anticipated when we created the Series 6 business case.

[Operator]

And we will now hear from Michael Weinstein with Credit Suisse.

[Maheep Mandloi, Credit Suisse]

Hi, this is Maheep Mandloi on behalf of Michael Weinstein. Thanks for taking the questions. Just building away from margins to the Safe Harbor. So the $315 [ph] million in the 5 gigawatt implies 8% pre-buy versus the 5%, I just wanted to understand what’s driving the higher pre-buy? And what would you be using of buying to satisfy the rule? And lastly, of that 5 gigawatt, how many projects or sites have been identified today? And as a follow-up, do you see any risk to the investment, if say, the democratic party is able to push for tax credit extensions? Thanks.

[Mark Widmar, Chief Executive Officer]

Let me just say one thing though and I’ll let Alex address the majority of that question is that, I don’t want to get into the specifics of how we’re looking to secure the Safe Harbor, I mean, there is a component of that is achieved through the physical work test and those would be – that would be related to what we have contracted today and that sites are pretty far along in their development.

But I don’t want to get into specifics of how we’re looking to do it, because I do think the concern I have a little bit, the more transparent we are, the more potential pressure that creates on procuring whatever we’re looking to source that with whether we use our own module in some case or whether we use third-party balance of system, whatever we choose to look to engage and procure the market, that’s going to give a signal that potentially would drive pricing higher than we would otherwise and we don’t want to be transparent in that regard. And so we’ve given ourselves many different options of how we would perfect the Safe Harbor buy to ensure we need the 5%, but I don’t want to be transparent with exactly our strategy and how we’re going to do that. So I’ll let Alex answer the rest.

[Alex Bradley, Chief Financial Officer]

Yes. And so to the other point that you raised, so yes, if you do the math on that, you may get higher than 5%, but bear in mind a couple of things, one is that, we may have storage on some of these sites and so you’re going to see different capital costs incur obviously you would blend in that as well. And we’re not going to give the exact split, but the majority of the spend is for un-contracted sites. So the spend on the contracted sites mostly in 2020, a little bit in 2021, the un-contracted bit in 2021 and beyond, but we continue to believe that will be pretty strong systems business. But if you look at that $325 million to $375 million that we have talked about, about $30 million of that is going to be associated with physical work that will fall into physical work test and the reminder of that, somewhere around $300 million and a little bit over $1 million will be used to precure equipment as Mark talked about we are not going to specify back to what will be the range across the module in the balance of system. The other thing just I want to make sure if we put it in perspective that again the 5% has to be against the fair market value basis that is used to determine the ITC and for those who may know the fair market value is more of an indication of DCF of cash flow generated relative to the actual capital acquisition cost. And generally, the fair market value is going to be higher than the capital acquisition cost. So therefore you have to Safe Harbor to a higher number. And then the other thing is you are going to want to give yourself some air, some buffer, alright, because you are not going to want to expose yourself, the value creation that you are going to have as a result of that, you may actually instead of using in the 5% threshold you many want to go to 6%, you may even want to go 7% just so you don’t have an issue when ultimate tax equity structuring gets done and comfort around the ability to perfect the Safe Harbor in 2022 or 2023.

The other one as it relates to your question around [indiscernible] get in and there is an extension of the ITC as Alex indicated we are just pulling forward working capital and we are largely going to focus on components at our commodity. So we are not looking to do components that have technology risk associated with it, such that if there was an ITC extension unforeseen at this point in time that did happen, those commodities are the commodities. And to some extent if you are in a rising commodity environment, you procured commodity to potentially lower value and if you had to wait and procure it in 2022. So I don’t see as the way we are looking at this point in time at our strategy that there is a risk associated with it plus we will feather this in between now and the end of ‘19. So we have got 12 months to engage and continue to asses and evaluate in an likely event, there was some type of extension to the tax credit. We can make a pivot at any point in time.

[Alex Bradley, Chief Financial Officer]

Yes, I think I will downsize is the worst case risk at the small cost of carry risk against that capital we outlaid, the upside I think is pretty significant as a compelling return against the spend on the assumption as there is not an extension of the ITC. So we feel it’s a prudent investment.

[Operator]

And our next question will come from Julien Dumoulin-Smith with Bank of America/Merrill Lynch.

[Julien Dumoulin-Smith, Bank of America/Merrill Lynch]

Hey good afternoon. Thanks for the time. Just a few quick questions if I can. First, just on lawsuit, can you give us a little bit more of a sense of the potential dollar exposure, I know it’s a little bit early without clarity on certain etcetera, but just give a little bit more of the scenarios? Then secondly with respect to the potential higher CapEx and decision to expand the S6 in 2020, should we think about that being ratably pretty equivalent to the initial Series 6 ramp or is it going to be at least a little bit cheaper as you think about the Malaysia piece of this? And then lastly just to clarify and Safe Harbor, the 5 gigawatts I know you wanted just some detail about this 5% or maybe little bit higher, but potentially could that flex higher or are you pretty firm about that 5 – only securing 5 gigawatts at this point in time given the dollars that you are putting down?

[Mark Widmar, Chief Executive Officer]

I will take the lawsuit thing. Julien, it’s way too early and we can’t really comment on anything at this point in time. This case has been around for a number of years. We have had a number of filings in our [indiscernible] and 8-Ks as it relates to the status of this case and it’s been continuous, involving and appeal after another appeal and now it’s at Supreme Court level and the primary issue that they are trying to get clarity around is the termination of loss causation and the standard which should be used to determine loss causation. And until that has been clarified, it’s really uncertain which way the case will go and in one case, if it’s certified or if it’s determined under one line of cases that was used, it could say that the minimal damages, the case it could be more significant. So, we don’t know until we get clarity around that. And it’s unclear exactly when the timing of that would happen. As Alex indicated, it’s – the Supreme Court has asked the Solicitor General to look at it. They will evaluate the case. They will make a recommendation to the Supreme Court. Even if they make a recommendation not to hear the case, the Supreme Court can still decide to hear the case. So it’s unclear until we have clarity on that, it’s really too premature to determine any potential outcome. The other thing we were trying to highlight again, because this case has been ongoing really since 2012 is that the insurance coverage is given the duration of which we have had to defend this case we could start to see a situation that either expenses to continue to support assuming we were to go to trial in 2019 or potential settlement could exceed our current insurance coverage. So, we will continue to provide more update as we move forward, but it’s really too early to give any indication of any range. As it relates to the Safe Harbor and our strategy around that, it easily could, Julien. And the other thing we had to continue to think about is that anything we do in 2019 preserves 30%. If we can make decisions in 2020 that preserves the 26%, right, for a longer period of time. So there is still compelling investment decision that will have to be made in 2020 and we will continue to assess what we are very fortunate though it have is very strong balance sheet and sufficient liquidity to make those decisions and we are going to look to try to capture the greatest value that we can. As I look forward between now and 2023, there is going to be a tremendous opportunity in the U.S. and I think we are going to see a pull forward of demand from 24, 25 and 26 type projects into 2023 and we want to be best positioned to take full advantage of that. So, I will let Alex answer the CapEx question.

[Alex Bradley, Chief Financial Officer]

Yes. On the CapEx, you can think about incremental factor is being similarly priced to the average of what we spent so far. We have talked about it having a CapEx profile about $0.30 per watt of additional CapEx. So, for a 1.2 gigawatt factory, which is the size we are building today throughout about $350 million. So if you took the insight from what we have in our existing sites that would take us up beyond the 6.6 gigawatts we talked about today essentially up to about 10 gigawatts you would have a little over another $1 billion of capital spend that isn’t factored into today’s plans.

[Operator]

And we will hear from Ben Kallo with Baird.

[Ben Kallo, Baird]

Hi, thanks for taking my questions guys. Just on the ramp cost, could you just talk to us, because I look back and look at your Analyst Day, I kind of square the cost for what you originally thought, I know you pulled forward some production ramps. Can you talk about how that lines up with the original thoughts on the ramp cost? And then my second question is on the capital allocation, you mentioned possibly returning capital to shareholders in evaluating that? Could you just talk about the different types of triggers throughout the year and just remind us kind of how comfortable you are and what kind of cash balance you are comfortable with going forward this year? Thanks.

[Mark Widmar, Chief Executive Officer]

Ben, on the ramp cost, unfortunately we will get back to you on that, because I don’t have in front of me the exact number that we have highlighted on the ramp cost. What I can tell you though is that we have been able to – in each successive factory we have actually accelerated a ramp. When you look at three factories we have now which compare our first startup for the plant to our first scale on plant to our first Vietnam plant, the ramp has actually accelerated in each one of those factories and the cost has come down. And we anticipate that to continue for our second factory in Vietnam. Relative to what we actually showed in the Analyst Day let us come back to you, Steve and I will look at those numbers and we will come back to you with the specifics.

[Alex Bradley, Chief Financial Officer]

And Ben talking a little bit about cash and capital, so as we discussed in multiple occasions, we have this approach of how we think about capital in terms of a waterfall through funding operations, manufacturing capacity expansion, development business, M&A buffer for industry volatility and at the end of that looking if we can’t see accretive uses of capital elsewhere looking at capital return. In terms of current cash position, when we are going to end 2018 with $2 billion to $2.2 billion in net cash when forecasting to end 2019 of $1.6 billion to $1.8 billion of net cash. In terms of other sorts of liquidity for us, we don’t have corporate level debt as you know. We have some project level debt and the majority of that is working capital that rolls off as underlying project is sold. And we have a revolving facility to $500 million. That facility actually requires maintaining a minimum cash balance of $400 million against it. That really exists with the legacy to when we first structured that revolver in 2009, the time when the credit profile around the solar states is very different. So, we maintain that and use it for LCs in managing jurisdictional cash. The LCs, we have some bilateral lines as well and the jurisdictional cash is less relevant now given tax reform, but still has some relevance given we maintain permanent investment status offshore for our cash. So we are looking at the revolver and seeing whether it makes sense to amend or dispense of that, but in actual fact, we only get about a $1 million of liquidity above that. So if you think about our total capital and we think of this as a strategic resource right, when we view it as a strategic resource and not clear that we are getting full value for it today, but given our experience, the volatility reflects factor and the financial track record of most people in this sector is clear, we need to take a conservative approach in liquidity. So if you think about how we look through capital return and the systems business and we historically and today support several $100 million of LCs and parent guarantee commitments in both the development EPC business. That’s normal course of business and we don’t expect liquidity events against that, but it will be prudent for us to maintain some coverage. We have talked just now about manufacturing capacity, if we were to expand up to the 10 gigawatts we talked about there could be another roughly $1 billion of capital needed there. Again, the capacity we are putting in is cash generative, but there maybe timing differences between when we want to spend that money and when we are generating cash. So we want to maintain that flexibility. The Safe Harbor is $325 million to $375 million this year. That’s something we would not have thought we would have needed earlier last year. So again having a flexibility to do that is important. So, we will continue to monitor this. We also want to make sure we maintain liquidity against either unknowns or current known risks and the class action we talked about is a good example right. We believe we have meritorious defense. We are going to continue to defend that rigorously, but ongoing defense costs would likely require OpEx increases and if there were resolution not in our favor, that will also require liquidity. So we are not in a position to declare anything today or to give you a firm number of what we believe the capital position needs to be. If we work for some capital, I don’t believe we are in a position to put in place recurring dividends. I think we will more likely look to the form of a share buyback authorization. But we will continue to evaluate that minimum liquidity requirement relative to the business parameters we know and the unknown risks and opportunities we will look over that as ‘19 progresses and keep giving updates throughout the year.

[Mark Widmar, Chief Executive Officer]

Hey, Ben, just Steve and I was – and Alex was taking we looked at the Analyst Day slide, and so I think when you look at the – there’s still performance what we showed similar to what we’re showing today, we show ramp and we show a start-up. If you look at the ramp costs between ‘19 and ‘20, $25 million a year or so, that means those numbers are pretty consistent with what we’re showing now. The number that has increased in 2019 significantly is start-up, I think the start-up number was around $25 million as well and that number is significantly higher. And there is two things that have changed on the startup side. The biggest one is, we’ve added a second factory in Perrysburg, right. So when we did the Analyst Day, we did not envision the expansion in Perrysburg, but we have made that commitment in the start-up, which Alex alluded to in his comments on Perrysburg is much higher than we would have anticipated and really not optimal. And part of the reason is, we’re – as we’re shutting down our Series 4 production in Perrysburg, we’re moving that workforce over to PVG2, our new plant in Perrysburg. And that’s going to require a much higher start-up expense because we’re bringing that workforce on sooner than we would otherwise. Now the reason we’re doing that is an experienced workforce that will also then enable us to ramp the factory much faster in 2020 than we would otherwise. So there is some inefficiency and again, Perrysburg is our highest cost jurisdiction, so just embedding that self will drive to higher startup cost. But the other one that is important to understand is we do have – when we included in our prepared remarks, we have $15 million – it’s actually slightly north of $15 million for a couple of vertical integrated projects that we’re looking to bring some of our current build material in house. We have not made a final decision on that, but as we look to make versus buy on a couple of those components, we believe we maybe able to drive cost down faster by bringing those components in house and the start-up expenses associated with that that’s a little bit north of $15 million.

As we continue to evaluate that and we are still negotiating with our suppliers, so it’s make versus buy, but that buy component will continue to move depending on how much leverage we have in terms of internal sourcing. If we can get to an outcome with our vendors that makes sense then we will maybe look to do something differently, if not, then we will vertically in-source and what we will see is a build material benefit and lower cost of our products, but there will be some initial start-up in 2019, which is about $15 million, it was not in the Analyst Day.

[Operator]

And our final question will come from Colin Rusch with Oppenheimer.

[Colin Rusch, Oppenheimer]

Thanks so much for squeezing me in. Can you talk a little bit about the trends on customer and geographic diversity in terms of the bookings and the opportunities that we’ve got in the next couple of years? And then beyond the – just what you’re talking about in terms of in-sourcing, are there other opportunities for material cost reduction with your supply chain as you look at scaling out to even higher volumes?

[Mark Widmar, Chief Executive Officer]

Yes. So on the geographic diversity, demand is robust globally kind of our issue is not so much ability to have visibility and line of sight to demand. It’s our ability to support and it’s not just that even in our traditional market segments, which is largely a utility scale, we are seeing quite bit of interest with, I will call it, smaller, larger C&I or smaller utility scale, so think of it as 1 megawatt to 20 megawatt type of opportunities. And a lot of interest especially with our Series 6 product and even now patients that we haven’t historically thought of like car parts and other things along those lines has actually even had one of our customer use some of our Series 4 product in the car ports and they are really interested in trying to get Series 6 for that type of application. So there is a lot of demand and it’s even not just diverse geographic jurisdictions, but also is segments and applications that we historically haven’t been involved with. It’s a matter of just lining that up with available supply. And our U.S. customers, some of our international customers are more interested in going out into ‘21 and ‘22 having those conversations, but a lot of them are still more focused on ‘19 and ‘20 of which we really don’t have a lot of supply. Now, last quarter, we broke down the 300 megawatts of our north of 1 gigawatts with international customers and I was very happy with that, but trying to get to we would like to see a longer term mix profile that it’s closer to 50:50 or maybe it’s 60:40, 60% U.S., 40% international, but near-term it’s going to be little bit more challenging, because there is such robust demand and great pricing. And until we add incremental capacity and trade the supply that enables us to look beyond what our current target is right now we maybe more focused – our profile maybe more concentrated in the U.S., but that’s not the long-term intent. We really want to continue to find ways to enable support international customers from that standpoint. Supply chain, sorry, I was trying to take the last question. We are looking at everything that we can, Colin. And that’s why we are looking at these variably integrated opportunities in. And the other ones that you mentioned is the other one that we are looking at right now is leveraging our purchasing power, alright. As we are going out into the market, given the volume ramp that we are seeing it in the not only north of 5, north of 6 potentially 7 and closer to 8. We are leveraging that forward-looking visibility and purchasing power to drive down cost even further and some of the things that we have done is as we referenced before we have tried to co-locate in Ohio, close to our Ohio facility at Glasgow, same thing in, as we put our facility in Vietnam is that there was at Glasgow [ph] that was close to the Vietnam facility which therefore drives down the cost to our suppliers in that regard and then capturing some of the value associated with that. So we have intense activities and motivation to work our supply chain, work with them, partner with them, but try to find a way that we can be as competitive as possible and driving the cost of our bill of materials. It really as we know cost wins today everyday and we have got to find it how do we make sure that everything we do is competitive and it’s not just the bill of materials that’s on our freight as well. One of the things we are working aggressively is how do we trading advantage on freight costs, how do we make sure that our freight costs in relative to crystalline silicon equal or if not better to them and our ability to drive down our freight costs for our modules. So, all aspects of the cost structure has to be continuously monitored in that regard.

[Operator]

And that does conclude our call for today. Thank you for your participation. You may now disconnect.