First Solar, Inc. Q3 FY2023 Earnings Call

Hello. Good afternoon, everyone, and welcome to First Solar's Third Quarter 2023 Earnings Call. This call is being webcast live on the Investors section of First Solar's website at investor.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 Richard Romero from First Solar Investor Relations. Richard, you may begin.

Good afternoon, and thank you for joining us. Today, the Company issued a press release announcing its second quarter 2023 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 provide a business update. Alex will discuss our financial results and provide updated guidance. 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 statement 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.

Thank you, Richard. Good afternoon, and thank you for joining us today. On our recent Analyst Day in September, we outlined our goal to exit this decade in a stronger position than we ever did. We believe the future belongs to thin film, and we described our long-term intent to be positioned to serve all addressable markets and commercialize the next generation of PV technology, balancing and optimizing across efficiency, energy, and cost in an environmentally and socially responsible way. This long-term aspiration aligns with our nearer-term growth, which is underpinned by our points of differentiation and solid market fundamentals, including continued strong demand for our products, proven manufacturing excellence, a uniquely advantaged technology platform, and crucially, a balanced business model focused on delivering value to our customers and our shareholders. This is our last earnings call. We have continued to make steady progress on this journey, and I will share some key highlights related to continued strong demand and ASPs, manufacturing operational excellence, and expansion. Beginning on slide three, we will continue to build on our backlog with 6.8 gigawatts of net bookings since our last earnings call at an ASP of 30 cents per watt, excluding India. This base ASP excludes adjusters applicable to approximately 70 percent of these bookings, which when applied with our aligned technology roadmap, may provide potential upside to the base ASP. These bookings bring our year-to-date net bookings to 27.8 gigawatts and our total backlog to 81.8 gigawatts. Our total pipeline of future bookings opportunity stands at 65.9 gigawatts, including 32.5 gigawatts of mid- to late-stage opportunities. As it relates to manufacturing, we produced 2.5 gigawatts of Series 6 modules in the third quarter with an average watt per module of 469, a top-end class of 475, and a manufacturing yield of 98 percent. Our third Ohio factory, which establishes the template for high-going Series 7 manufacturing, continues to ramp, demonstrating the manufacturing production capability of up to 15,000 modules per day, which is approximately 97 percent of nameplate throughput. The factory produced a total of 565 megawatts in Q3. Total year-to-date production of Series 7 modules in the U.S. has surpassed one gigawatt. As noted on our Analyst Day, the factory recently demonstrated a top module wattage produced of 550 watts as part of a limited production run. While still undergoing commercial qualification testing, this implies a record production module efficiency of 19.7 percent. Our India plant started production in Q3 and is continuing to ramp, recently demonstrating a manufacturing production capability of approximately 12,000 modules per day, which is approximately 77 percent of nameplate throughput. The factory produced a total of 154 megawatts in Q3 and recently demonstrated a top module wattage produced of 535 watts. In terms of technology, our Series 6 plus bifacial modules completed rigorous field and laboratory testing. We recently converted our first Series 6 plus plants in Perrysburg, Ohio, to commercially produce the world's first bifacial solar panel, utilizing an advanced thin film semiconductor. The technology features an innovative, transparent back contact pioneered by First Solar's research and development team, which in addition to enabling bifacial energy gain, allows infrared wavelengths of light to pass through rather than be absorbed as heat and is expected to lower the operational temperature of the bifacial module and result in higher specific energy yield. Upon successful demonstration of operational metrics in high-value manufacturing, such as yield and throughput, we plan to convert more plants in the future, which will enable us to capture incremental ASP through our existing contractual adjusters. Turning to slide four, our focus on delivering value extends to our manufacturing expansion strategy, and we are making tangible progress towards achieving our forecasted 25 gigawatts of global nameplate capacity by 2026. Construction of our India facility is completed, and production has commenced. Commercial shipments are expected to begin once the factory receives the Bureau of India Standards certification from the Indian government, which is expected by year-end. In September, we mobilized on our new Louisiana manufacturing facility, our fifth fully vertically integrated factory in the United States. At a ceremony attended by the governor of Louisiana, we set in motion the work expected to deliver 3.5 gigawatts of annual nameplate capacity, which is anticipated to commence operation at the end of 2025. When completed, we expect to have approximately 14 gigawatts of annual nameplate capacity in the United States, further enhancing our ability to serve the market with domestically made modules. Meanwhile, we continue to make steady progress on the construction of our new Alabama facility, which is expected to commence operation in the second half of 2024, and our Ohio manufacturing expansion, which is projected to commence operation in the first half of 2024. Additionally, our new R&D Innovation Center and our first perovskite development line, announced at Analyst Day, are also on track, representing an expected combined investment of $450 million. The perovskite development line and R&D Center are expected to commence operation in the first half of 2024 and reflect our determination to lead the industry in developing the next generation of PV technologies, optimizing across efficiency, energy, and cost. Crucially, as our manufacturing footprint continues to grow, so does our supply chain. In the U.S., we recently expanded our agreement with Vitro Architectural Glass, which is investing in upgrading existing facilities in the United States to produce glass for our solar panels. The expanded capacity commitment from Vitro to First Solar is expected to commence production in the first quarter of 2026. Today, First Solar is one of the largest consumers of float glass in the United States. As PV manufacturing continues to scale in the U.S. and a premium is placed on domestically produced components, including glass, our early work to build a resilient domestic supply chain, which began in 2019, gives us a significant head start over the competition. Similarly, we expect Omco Solar to manufacture and supply Series 7 module back rails through a new facility in Alabama. This reflects our efforts to de-risk our supply chain with strategic localization. Omco only uses American-made steel, which aligns with our intent to maximize the domestic economic value created by our U.S. manufacturing footprint. Similarly, our high facility is also served by a steel value chain that is located within a 100-mile radius of our factories. Before handing the call over to Alex, I would like to discuss our policy environment. In the United States, with regards to the Inflation Reduction Act, we continue to await guidance from the Department of Treasury on the Section 45X manufacturing tax credits. We also remain engaged with the administration and continue to work with our customers to ensure that the IRA's domestic content bonus guideline supports the Act's intent to sustainably grow U.S. solar manufacturing and its supply chains. As we have previously noted, we share our commitment to the current guidance with the administration and are working with our customers to enable their ability to benefit from the bonus credit for using U.S.-made content. We are appreciative of the work done by the Biden administration to provide a solid legislative foundation for domestic solar manufacturing. The IRA has supplied a catalyst for growth, and our goal is to leverage it to help create a position of strength for the country both now and after the term of the incentives. Beyond the IRA, we are also aware of new anti-dumping and countervailing duty petitions filed against importers of aluminum extrusions from 15 countries. Consistent with our views on fair trade and the importance of conforming with rules governing trade issues, we will comply with any request for information from the United States Department of Commerce and the International Trade Commission while we work to understand the potential implications. Moving abroad, we remain engaged with policymakers across Europe as the bloc attempts to tackle serious challenges to its solar manufacturing ambitions. For example, Chinese module inventory in Europe, stored in warehouses across the region and estimated by analysts to reach 100 gigawatts by the end of the year, is reportedly being sold at prices below its cost to manufacture. This alleged dumping behavior, driven by overcapacity in the Chinese silicon industry that has decimated international competition for the past decade, represents a serious threat to non-Chinese manufacturing and to ambitions of diversifying solar supply chains away from the dependency on China. It also represents a policy threat, potentially undermining the political willingness to deliver the comprehensive trade and industrial legislative solutions necessary to both level the playing field and incentivize domestic manufacturing. We continue to advocate for comprehensive legislation to safeguard any domestic manufacturing ambitions. Our view is that industrial manufacturing ambitions, our view is that industrial policy-related CapEx benefits alone are insufficient, and that absent sufficient trade policy support to ensure a level playing field, Europe will find it challenging to achieve what the U.S. and India have been able to do in a relatively short period of time. I'll now turn the call over to Alex, who will discuss our bookings, pipeline, and third-party results.

Thanks, Mark. Beginning on slide five, as of December 31, 2022, our contracted backlog totaled 61.4 gigawatts, with an aggregate value of $17.7 billion. To September 30, 2023, we entered into an additional 23.6 gigawatts of contracts, and recognized 7.4 gigawatts of volume sold, resulting in a total contracted backlog of 77.6 gigawatts, with an aggregate value of $23 billion, which equates to approximately 29.6 cents per watt. Since the end of the third quarter to date, we've entered into an additional 4.3 gigawatts of contracts, contributing to our record total backlog of 81.8 gigawatts. Including our backlog since the previous earnings call, our contracts are approximately one gigawatt or more, with returning customer Long Road Energy, and new customers, including a new IPP, and an asset manager with multiple companies in its portfolio. Additionally, we have received full security against 141 megawatts of previously signed contracts in India, which now move to these volumes from the contracted subject conditions precedent grouping within our future opportunities pipeline to our bookings backlog. As noted on this day, while the ASPs associated with these India bookings are lower than those associated with the 6.6 gigawatts of US bookings since the prior earnings call, gross margin profile, excluding the 45x benefit, is comparable to the fleet average given the lower production costs in our Chennai facility. Since the announcement of the IRA, we've amended certain existing contracts to provide US manufactured products, as well as to supply domestically produced Series 7 modules in place of Series 6. Consequently, over the past five quarters, to the end of Q3 2023, across approximately 11 gigawatts, we've increased our contracted revenue by approximately $354 million, an increase of $42 million from the prior earnings call. As we previously addressed, a substantial portion of our overall backlog includes the opportunity to increase base ASP through the application of adjusters, if we're able to realize achievements within our technology roadmap, at the required time of delivery of the product. As of the end of the third quarter, we had approximately 40.3 gigawatts of contracted volume with these adjusters, which if fully realized, could result in additional revenue up to approximately $0.4 billion, or approximately $0.01 per watt, the majority of which we recognize between 2025 and 2027. As previously discussed, this amount does not include potential adjustments, which are generally applicable to the total contracted backlog. Both the ultimate-bin produced and delivered to the customer, which may adjust the ASP under the sales contract upwards or downwards, and for increased sales rate or applicable aluminum or steel commodity price changes. Our contracted backlog extends into 2030, and excluding India, we are sold out through 2026. Note, a total of approximately 1.5 gigawatts of production from our India facility is expected to be used to support US deliveries in 2024 and 2025. As reflected on slide six, our pipeline potential bookings remains robust. Total bookings opportunities are 65.9 gigawatts, a decrease of approximately 12.4 gigawatts as of the previous quarter. Our mid- to late-stage opportunities decreased by approximately 16 gigawatts to 32.5 gigawatts, and includes 27.1 gigawatts in North America, 3.8 gigawatts in India, 1.3 gigawatts in the EU, and 0.3 gigawatts across all other geographies. Decreases in our total mid- to late-stage pipeline in Q2 to Q3 result both from converting certain opportunities to bookings as well as a removal of certain other opportunities given our sold-out position and diminished available supply. As we previously stated, given this diminished available supply, the long-dated timeframe into which we are now selling, and aligning customer project visibility with our balanced approach to ASPs, field security, and other key contraction terms, we would expect to see a reduction in volume in upcoming quarters. We will continue to forward contract with customers who prioritize long-term relationships and value our differentiation, and given the strength and duration of our contracted backlog, we will be strategic and selective in our approach to future contracts. Included within our mid- to late-stage pipeline are 5.1 gigawatts of opportunities that are contracted subject to conditions present, which includes 1.7 gigawatts in India. Given the shorter timeframe between contracting and product delivery in India relative to other markets, we would not expect to see a multi-year contract commitment to occur in the U.S. As a reminder, signed contracts in India will not be recognized as bookings until we have received full security against the Arctic. Next slide, 7, I'll cover our financial results for the third quarter. Net sales in the third quarter were $801 million, a decrease of $10 million compared to the second quarter. The decrease in net sales was primarily driven by lower non-module revenue associated with project earn-outs from our former systems business, as well as within the module segment, a slight reduction in volume sold, partially offset by an increase in ASPs as we continue to see favorable pricing trends. Gross margin was 47% in the third quarter, compared to 38% in the second quarter. This increase was primarily driven by higher module ASPs, lower sales rate costs, and higher volumes of modules produced and sold in the U.S., resulting in additional credits from the Inflation Reduction Act. As previously mentioned, based on our differentiated vertically integrated manufacturing model, the current form factor of our modules, we expect to qualify for an IRA credit of approximately $0.17 per watt for each module produced in the U.S. and sold to a third party, which is recognized as a reduction to cost of sales in the period of sale. During the third quarter, we recognized $205 million of such credits, compared to $155 million in the second quarter. We encourage you to review the safe harbor statements contained in today's press release and presentation, the risks related to our receiving the full amount of the benefits we believe we are entitled to under the IRA. The reduction in our sales rate costs during the quarter reflected improved ocean and land rates, along with a beneficial domestic versus international mix of volume sold. Lower sales rate costs reduced gross margin by 7 percentage points during the third quarter, and by 8 percentage points in the second quarter. Ramp costs reduced gross margin by 3 percentage points during the third quarter, and by 4 percentage points during the second quarter. Our year-to-date ramp costs are primarily attributed to our Series 7 factory in Ohio, which is expected to reach its initial target operating capacity later this year, and our new Series 7 factory in India, which commenced production during the quarter. S&A and R&D expenses total $91 million in the third quarter, an increase of $8 million compared to the second quarter. This increase is primarily driven by expected credit losses associated with our higher accounts receivable balance, additional investments in our R&D capabilities, and costs related to the implementation and support of our new global electrified resource plan. Production start-up expense, which is included in the operating expenses, was $12 million in the third quarter, a decrease of approximately $11 million compared to the second quarter. This decrease was attributable to the start-up production in our factory in India, partially offset by certain start-up activities for our new Series 7 factory in Alabama. Our third quarter operating results did not include any significant non-module activities. However, the year-to-date operating loss impact for the legacy systems business related activities remains at approximately $22 million. Our third quarter operating income was $273 million, which included depreciation, amortization, and accretion of $78 million, ramp costs of $25 million, production start-up expense of $12 million, and share-based compensation expense of $8 million. We recorded tax expense of $22 million in the third quarter, and tax expense of $18 million in the second quarter, primarily driven by higher pre-tax income. A combination of the aforementioned items led to a third quarter diluted earnings per share of $2.50 compared to $1.59 in the second quarter. Next on the slide, eight, I will discuss the expense items and summary cash flow commission. Our cash, cash equivalents, restricted cash, restricted cash equivalents, and marketable securities ended the quarter at $1.8 billion, compared to $1.9 billion at the end of the prior quarter. This decrease was primarily driven by capital expenditures associated with our new facilities in Ohio, Alabama, and India, along with our higher accounts receivable balance, partially offset by advanced payments received from future module sales. Total debt at the end of the third quarter was $499 million, an increase of $62 million since the second quarter, as a result of the final loan drawdown and credit facility for our factory in India. Our net cash position decreased by approximately $0.2 billion to $1.3 billion as a result of the aforementioned factors. Cash flows from operations were $165 million in the third quarter. Global liquidity and the strength of our balance sheet remain one of our key differentiating factors. However, as discussed on our Analyst Day, the majority of our cash sits offshore, while the majority of our forecasted future CapEx spend between 2024 and 2026 is in the United States. As we invest significantly in U.S. manufacturing ahead of any IRA cash proceeds, we continue to evaluate options to optimally balance this expected temporary jurisdictional cash imbalance, which includes cash repatriation, the use of our existing undrawn revolving credit facility, or other sources of capital. While we expect our $1 billion of revolving capacity to provide sufficient liquidity, we continue to evaluate other options to optimize the cost of capital for any French financing. On slide nine, our guidance updates, our volume sold and net sales guidance remains unchanged. Within gross margin, we are reducing the high end of our forecasted ramp under the utilization expenses by $10 million, between $110 and $120 million while narrowing the range of our Section 45X tax credit guidance by $10 million, both the low and high end, between $670 and $700 million. Given their size, these combined changes do not impact our guided gross margin range of $1.2 to $1.3 billion. We've reduced our production start-up expenses guidance to $75 to $85 million, which implies operating expenses guidance of $440 to $470 million. Combining these changes provides some resiliency to the low end of both the operating income guidance range, which is updated to $770 to $870 million, and the earnings per share guidance range, which is updated to $7.20 to $8. Net cash and capital expenses guidance remains unchanged. Turning to slide 10, I will summarize the key messages from today's call. Demand continues to be robust, with 27.8 gigawatts of net bookings year-to-date, including 6.8 gigawatts of net bookings since our last earnings call, and an average ASP of 30 cents per watt, including India. And before the application adjusters were applicable, leading to a record contracted backlog of 81.8 gigawatts. Our continued focus on manufacturing technology excellence resulted in record quarterly production of 3.2 gigawatts. Our India manufacturing facility commenced production, and our Alabama, Louisiana, and Ohio manufacturing expansions remain on schedule. Financially, we're on track for $2.50 per diluted share, and we ended the quarter with a gross balance of $1.8 billion, or $1.3 billion net of debt. We maintain fully-expensed 2023 revenue guidance and raise the midpoint of our EPS guidance from $7.50 to $7.60. With that, we conclude our prepared remarks and open the call for questions.

Your first question comes from the line of Philip Shen from Roth MKM. Please go ahead. Philip, your line is open.

Hey, guys. Thanks for taking the questions, and congrats on the strong bookings that appear to be at strong pricing. Congrats on the strong bookings, what appears to be strong pricing. Mark, can you talk through the pricing at $0.30 a watt that's without India? And I think the prior quarter, there were some nuances around a contract without freight. So if you adjusted that where you typically include freight, was your prior pricing kind of closer to $0.31? So you guys are sitting close to $0.30 this quarter with maybe a bit of a drop, but really compared to the crystalline silicon price collapse, it looks like you're holding price pretty well. And then looking ahead, I think you guys said you may be selective and strategic with bookings. Should we expect things to slow down from here and maybe fewer bookings in general coming up in this full quarter here in Q4 and maybe in Q1 as well, especially since U.S. LPA module you have compliance module pricing has come down so much there? I'm just curious what you expect ahead there as well.

Yes. From a branding perspective, if you examine the bookings for this quarter extending into 2029, you'll notice that much of it is actually focused on 2029. We're looking further ahead, which influences our base pricing and the effect of the additional costs. As we mentioned, the $0.30, excluding India, does not factor in those additional costs, and 70% of the volume comprises them. There's a significant opportunity for us to enhance our position, especially regarding the long-term degradation rate influenced by temperature. As noted in the call, we're beginning our initial production in Ohio. When considering the effect on the average selling price, including the benefits of the additional costs alongside our technology roadmap, you would see an increase of about $0.02 in ASPs. When you adjust for this and compare to last quarter, while booking periods extend further, I believe the pricing remains stable from quarter to quarter. It's true that last quarter, we had a sizable deal that did not factor in sales rate, which slightly affected the average ASP. Overall, pricing remains stable, and we're encouraged by our ability to forecast further out while still achieving appealing pricing, even amidst many changes in a dynamic environment over the past couple of months. Regarding our approach, we will maintain our discipline. We remain supply constrained, with a roadmap aimed at reaching 25 gigawatts. We're starting to see 27 filling up nicely and are beginning to secure more points for 28 and 29, while we have also touched 30 in previous deals. Engaging with customers must involve mutual agreements that make sense for us, not just in terms of ASP but also concerning security and overall conditions relevant to domestic content. This will be a balancing act in negotiations. We'll observe how the market responds, particularly regarding future bookings; some customers may hesitate to commit to the longer term, but we'll see how things evolve. It is also possible that bookings could decline slightly over the next several quarters.

Your next question comes from the line of Julien Dumoulin-Smith from Bank of America. Please go ahead.

Hey, guys. It's Alex on for Julian. Just a follow-up if I can to that, Mark. When you think about where you guys are booking, and I'll say this like you guys used to be in the development game as well. So I think you obviously understand the lead times on these projects. I mean, how much is that mid- to late-stage compression a function of just, listen, there's a lot of uncertainty as far as timing of interconnects, permitting, etc.? And looking out in 2028, it's sort of hard to say which projects will be first versus second versus third. Or is this more that the market is kind of getting back to some level of normalcy as far as supply and demand in modules and buyers are just electing to it? I guess sort of parse that for us relative to it just being really long dated as opposed to a sort of a shift in buyer sentiment or market conditions, if you will?

I don't see it as a huge shift in our customers' sentiment as they think about their realization against their development pipeline. Look, there are challenges as you indicated, permitting interconnection and what have you. But I think they all still are very bullish about their ability to realize their contracted pipeline and secure off-take agreements. The issue, I think, is around when do you actually if we're contracting for module deliveries in 2029, and we're asking for security. Clearly, a project is not in a condition at that point in time where they would be able to get financing put in place. So when you're talking about corporate liquidity capacity that's going to have to be used in order to provide the security, whether it's a parent guarantee and LC or actual cash. As you know, the project has to be much further along as it relates to financing debt and structure of tax equity before that liquidity is brought into the mix at the project level. So I think part of it is wanting to have the certainty of the delivery, but balancing that with capacity from a security standpoint that we're requiring on our contracts, and it's just a matter of finding a good balance that can work. Parent guarantees for certain entities can work, but we want to make sure they're creditworthy parent guarantees and guarantees that those guarantees are issued against that sometimes for some of our customers becomes a little bit more challenging. So you kind of got this balance of wanting certainty, wanting to engage, clearly want to partner with First Solar, and they also know that we're a loyal supplier to, especially our partners that have been with us for an extended period of time. But then also balancing their near-term liquidity constraints to the extent that they have when do they want to undertake contracts. So I don't see it so much of the sentiment to realize against the development pipeline. I just think it's you're going out to the horizon right now that people are maybe not as ready yet to commit capital and commit to the liquidity that we need to get comfortable with around security for module agreement.

Two things I might note. One is, at the Analyst Day, we talked around the fact that we actually over-allocate in the near term. And we do that deliberately because we tend to see projects move out to the right, that gives us some comfort. The other reason we do that is a lot of our recent bookings have been framework agreements with customers, whereby they don't necessarily have a specific project allocating the modules they're buying from us. They just know they're going to need that total volume over a period of time. And those frameworks can be more challenging to plan for because there is often some flexibility in timing there. but also shows that customers and very long-dated bookings are willing to buy without necessarily knowing exactly where the products go in because they value that certainty, and they know that over time, they'll find a home for it. So we've been seeing a lot more of behavior, which runs a little counter to your question, but we're seeing people looking out at times and they don't necessarily know exactly where it's going, but they're still willing to make that commitment because of the value I'm doing so. But as Mark said, the further we get out, the fact that we're now looking out into 2028, '29, it becomes part of people put meaningful deposits down and there's just less visibility on the framework side. That's why we talked about potentially seeing bookings slower.

Your next question comes from the line of Brian Lee from Goldman Sachs. Please go ahead.

Hi, thanks for taking the question. This is Grace filling in for Brian. My question is about competition. One of your competitors in the crystalline sector recently announced a 5 gigawatt expansion, which seems to be the first indication of differentiation from China in the U.S. Can you share your insights on the cost structure of overseas competitors?

There's a lot happening, especially with a recent announcement from one of our competitors regarding sales in the U.S. Other companies are also making moves in the U.S. market, such as Myer Burger, which has committed to selling here. However, it's important to clarify that while they are vertically integrated, they do not encompass the entire polysilicon production process. They are assembling modules and manufacturing cells, but wafer production and ingot manufacturing are still not taking place in the U.S. Additionally, the source of their polysilicon is uncertain, possibly coming from U.S. manufacturers or from Europe or Korea, making it an unequal comparison. Looking at their announcements, their sales figure is approximately $800 million, plus an additional $200 million to $300 million for their module, totaling around $1.1 billion, which is somewhat comparable to our fully integrated model. However, from a competitive perspective, the most revealing statistic is their headcount; they have 2,700 employees for just cell and module production for 4 or 5 gigawatts, whereas we are planning for 14 gigawatts of full vertical integration. Our entire workforce for that capacity will be comparable to their headcount, meaning they operate at about 2.5 times our employee count. This difference potentially adds around $0.02 to $0.025 to their cost per watt due to higher U.S. labor costs. Moreover, they lack a local supply chain, whereas we have established ours effectively. Our components for the Series 7 product are primarily sourced in the U.S. While their factory may require at least one major component sourced from outside the U.S., such as glass—which currently isn't manufactured domestically and would incur high shipping costs—they also need to consider potential duties on imported materials. In contrast, our Series 7 product is made from domestically sourced steel and does not face that exposure. I am confident in our position, and as we have stated before, we seek a level playing field for competition under the same policy environment. With this, I firmly believe we have a significant cost advantage over any other U.S. manufacturer for the reasons I've outlined. We are ready to compete with anyone choosing to manufacture in the U.S. and welcome that competition. Success under the IRA hinges on creating a diversified supply chain incorporating various technologies, whether it's prism silicon films, solar technologies, or perovskites. To achieve long-term energy independence and position the U.S. as a technology leader, we need more manufacturing innovation and diverse technologies.

Your next question comes from the line of Joseph Osha from Guggenheim Partners. Please go ahead.

Thank you and hello, everyone. Happy Halloween. Following on the previous question, assuming that most of what we see in the U.S. is going to be modules sourced with domestic cells, but almost certainly overseas wafer and poly. Based on what you see right now, can you see those suppliers managing to meet domestic content requirements under the IRA? And if so, just why we're not, I'm curious as to what your thinking is on that?

Currently, we understand the supply chain and the availability of domestically sourced components identified under the IRA domestic content guidance. The main component we believe will significantly contribute to domestic content is the cell. According to the most recent announcement, production should begin by the end of 2025, which means these cells will be available for production and shipments and eventually incorporated into models for installation in 2026. The IRA requirements for that year start at 40% domestic content, increasing to 55%. By the time the benefits of domestic content can be realized, requirements will already be at a higher threshold than they are now. Our analysis, focusing on the cell and accounting for direct material and labor costs in crystalline silicon, suggests that a module using only domestically sourced cells may struggle to meet the project level requirements for the domestic content bonus. However, Series 7, which constitutes the majority of our 14 gigawatts of magnetic production, is entirely sourced domestically, qualifying it as a domestic product and giving it a material advantage for obtaining the domestic content bonus at the project level compared to a crystalline silicon module with a domestically sourced cell.

Your next question comes from the line of Vikram Bagri from Citi. Please go ahead. Vikram, your line is live.

Sorry about that. Good evening, everyone. I wanted to ask about capital allocation. At the Analyst Day, we understood that there is some downside to tech CapEx by implementing some processes at the supplier level. Any updates to share there? Also, Alex, you mentioned that repatriating cash, it sounds like, is not the most efficient path to fund CapEx in the U.S. So what the cost of repatriation is. And staying on the same topic, understand that funding buybacks with cash is not on the table yet. I was wondering if repatriating the cash longer term can fund buybacks in the long term? And then finally, I was wondering if common equity is still off the table completely.

I'll take the first one and then Alex take all the rest of them. So yes, we are still working through with our supplier to enable various coding and capabilities that would result in us not having to make substantial capital investments related to our technology. Testing is ongoing. What I'll say is the early indications, a long way still to go. I want to make sure it's very clear. It's a long way still to go, but early indications on what we've seen so far is very promising that we will be able to find a way to provide or to have a supplier provide the coatings to the glass without us having to do on our own. Now look, there's some trade-offs with that such as the CapEx balance is also the opportunity to further optimize the buffer layer, which is what they're putting on to capture better performance at the semiconductor level. So we'll have to continue to assess respective trade-offs. But I would say, at least as of right now, really early innings. I want to continue to stress that there's a pretty positive indication of their capabilities in that regard.

If you consider our capital expenditures, we previously outlined a plan for 2024, 2025, and 2026, estimating spending between $3.5 billion and $4 billion. As Mark mentioned, there are initial signs that we might be able to reduce some technology-related capital expenditures. However, for the near term, most of our spending in 2024 will be focused on capacity expansion, R&D facility maintenance, and sustaining capital expenditures, so the guidance we provided, ranging from 1.6% to 1.9% for next year, does not heavily incorporate technology spending. Looking further into the coming years, there will be more technology-related expenses, and any potential reductions would likely occur towards the end of 2025 and into 2026. We will maintain a significant capital expenditure program for next year. Referring back to the 2017 tax reform, it resulted in a one-time transition tax that was akin to paying federal taxes for repatriating funds, which means most federal tax obligations are settled. However, there may still be state and local tax considerations when repatriating funds. Currently, we intend to permanently reinvest capital offshore. If we were to reverse that decision and bring capital back, the tax implications would depend on our profit and loss situation at that time, which I haven't quantified yet, but it could involve some significant tax considerations. Regarding funding needs, right now we're focusing on securing transition capital to bridge the gap between our substantial upfront investments and the cash inflow associated with tax credits. As mentioned during our Analyst Day, if we had immediate cash available when we recognized the tax benefit on our profit and loss statement, we wouldn't face potential challenges with timing and jurisdictional mix. However, I currently do not foresee a need for equity financing. In response to your question about buybacks, we haven't considered that yet and believe we're still quite far from needing to address that. We are embarking on significant capital expenditures in the coming years, and as such, we will evaluate buybacks when the time is right, but for now, our focus is on investing internally.

Your next question comes from the line of Colin Rusch from Oppenheimer & Company. Please go ahead.

Thanks so much, guys. Can you talk about how much finished goods inventory you exited the quarter with and where you're at right now in terms of the nameplate run rate in India?

So let me sure, Colin. So you want to know the enterprise-wise finished good inventory amount? Is that question not just India, right?

Yes. That's for the whole company and then understand where you're at in terms of the production run rate in India right now?

Yes, the company ended the quarter with over 3 gigawatts in inventory. Currently, we've produced about 150 megawatts in India, but that inventory is not yet ready for shipping due to certification delays. This has contributed to the increase in inventory. However, the inventory levels align with our anticipated shipments for the fourth quarter. In India, we've demonstrated close to 80% of our capacity, and we're currently operating at just under 70% of that. This is a significant achievement considering we initiated operations at that factory in July, and within three months, we were producing 10,000 modules a day. This shows a marked improvement and confirms our ability to produce finished modules at that rate. It's worth noting that the startup was largely a cold start due to permitting restrictions, and we couldn't fully activate the tools until we started the integrator fund, which allowed us to transition into our regular production process quickly. The results have been promising, and we hope this will serve as a positive indicator for our upcoming factories in Alabama and Louisiana. Our aim is always to commence operations at these new factories more efficiently than previous ones. The transition from Ohio to India has been encouraging, but there is still much work ahead, and I’m pleased with the factory's current performance.

Your next question comes from the line of Ben Kallo from Baird. Please go ahead.

Hey, Mark and Alex. Just on that note. I guess the question is two-fold. What do we think about your customers breaking contracts as that happens because soon going to open up a factory in Indiana or something like that? And how do we know that's not a risk? And number two, what you said there is the speed to time of your factories, I think, is getting better as they get more automated. And how does that factor into your ROIC or however you look at?

Ben, we recently completed a deal this quarter that should be announced in a press release this week, adding another 500 megawatts to an existing partnership, bringing our total with this partner to over 3 gigawatts. Our relationship has fostered a mutual understanding of the value First Solar can provide, particularly in delivering certainty that clients seek to mitigate risks associated with their projects. These multiyear investments involve substantial capital and are evolving with increased costs due to storage integration and potential hydrogen solutions essential for project success. Our partners primarily seek certainty and competitive technology at favorable prices, which helps them meet their commitments to their boards and shareholders. First Solar is uniquely equipped to fulfill these needs, especially with our Series 7 module, which we believe is the highest domestically produced module available. There is uncertainty regarding some production timelines, and a part of the output will be used for self-consumption by our partner's development arm, similar to our Ohio factory. This creates a different mindset for some partners who may hesitate to purchase technology from a competing developer that could potentially take market share from them. Moreover, ongoing changes, such as potential tariffs on aluminum from China and Southeast Asia, add to the risk profile that partners must consider. However, with First Solar, they can significantly minimize these risks while getting a high-quality product at a great price. While we have contractual penalties in place, the benefits of staying with us outweigh the risks of breaking contracts for uncertain gains. It doesn’t seem prudent to jeopardize their domestic content qualification just for minor, uncertain benefits. Regarding factories, every new factory we start up enhances our return on invested capital, particularly in the U.S. manufacturing sector. Accelerating product delivery improves our returns, and we look to factor in any potential new factories that could shorten the timeline from announcement to high-volume production, making our payback scenario more appealing.

Just about a couple of numbers around the 7 Asian fees. We center Analyst Day that about 14% of the megawatts in our backlog at that point was subject to a termination convenience close. If you look at that, I mean 86% of our alt scenario development that put themselves in a very difficult position because they have ongoing to see an contractual breach, which will make it very hard for them to seek financing and tax equity for a project going forward. But for the vast majority of our backlog, there is no ability to terminate for convenience. For those contracts that we do have that clause, which typically is when we have larger long-dated contracts, and we have a small portion of that contracts where we have subject some of the megawatts to nation book convenience. We then have an agreed fee typically up to 20%, which we look to collect the idea being that we could then resell those modules and be at least may follow on that transaction. So just to give you some color on the numbers.

Our final question comes from Andrew Percoco from Morgan Stanley. Please go ahead.

Thank you so much for taking my question. Mark, you sort of answered my question already, but I kind of just want to dive into the cost of capital environment. Obviously, that has an impact on the market or the perceived economics of renewable. I'm just wondering if you're seeing any developers or customers that maybe haven't been big solar customers historically that are maybe turning to your technology because they see your technology and your supply chain as more bankable than someone else? I'm curious if UFLPA, ADCVD combined with a more expensive cost of capital environment. I'm just wondering if that's becoming a bigger competitive advantage than it maybe was a year or two ago? Thank you.

I think Alex actually referenced it in his section, but if you look at our bookings this last quarter, we highlighted three large contracts that were over a gigawatt at that total booking time. One of them is a return customer that we made an announcement on with Long Road Energy. I think we made that right around September. But then we announced there were two other new customers. One is an IPP and another is effectively an asset management entity with a portfolio company and multiple developers both new customers. We're very happy with those in the first step of our journey of developing a deeper partnership with those counterparties. Look, they've come to First Solar for understanding of the unique value proposition and what we can provide. Unique value proposition and what we can provide. One of them, in particular, I know who would have liked to have gotten on First Solar's books earlier. We just didn't have capacity. And so now when they look forward and they see there is some supply you get out of '27, 28, '29. They want to secure some of that supply. They were lumped even on the books and '24, '25 and '26, in particular, we did it on supply. So yes, I do think that the environment that we're in right now and First Solar capabilities to proposition, I think, are more compelling and are driving new customers into our portfolio and our overall contracted backlog, which is now north of 80 gigawatts. I mean just that can reflect on that number. I mean that's a huge multi-year contracted backlog and commitments with dozens of different partners that uniquely understand First Solar and understand the value proposition that we can trade that enable the success of our business model.

And this concludes today's conference call. Thank you for your participation, and you may now disconnect.