使用警語:中文譯文來源為 Google 翻譯,僅供參考,實際內容請以英文原文為主
Operator
Good afternoon, everyone, and welcome to First Solar's Fourth Quarter 2020 Earnings and 2021 Guidance Call. This call is being webcast live on the Investors section of First Solar's website at investor.firstsolar.com.
(Operator Instructions) As a reminder, today's call is being recorded.
I would now like to turn the call over to Mitch Ennis from First Solar's Investor Relations. Ms. Ennis, you may begin.
Mitchell Ennis - Manager of IR
Thank you. Good afternoon, everyone, and thank you for joining us. Today, the company issued a press release announcing its fourth quarter and full year 2020 financial results as well as its guidance for 2021. 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 update. Alex will then discuss our financial results for the fourth quarter and full year 2020. Following these remarks, Mark will provide a business and strategy outlook. Alex will then discuss our financial guidance for 2021. Following the remarks, we'll 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, including among other risks and uncertainties, the severity and duration of the effects of the COVID-19 pandemic. 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 R. Widmar - CEO & Director
Thank you, Mitch. Good afternoon, and thank you for joining us today. I would like to start by expressing my gratitude to the entire First Solar team for their hard work and perseverance throughout 2020. Although 2020 was a very challenging year, I'm proud of the way our team responded with their ongoing commitment to health and safety, delivering value to our customers and achieving our objectives in this unprecedented year.
While Alex will provide a more comprehensive overview of our 2020 financial results, I would like to first note our full year EPS results of $3.73. This result came within but towards the low end of the guidance range we provided at the time of our third quarter earnings call, largely due to the volume and timing of our Sun Streams 2 project sale. Despite this timing impact, continued intense competition across the crystalline PV supply chain and unforeseen challenges related to the pandemic, we are very pleased with our financial and operational results in 2020.
Turning to Slide 3. I will discuss some of our key 2020 accomplishments. Firstly, our vertically integrated manufacturing process, diversified supply chain and differentiated CadTel technology enabled us to mitigate potential disruptions to our manufacturing operations from the pandemic. Accordingly, we produced 5.9 gigawatts of Series 6 and exited the year with a top production bin of 445 watts.
Secondly, driven by continued strong manufacturing execution, in Q4, we achieved a year-on-year 10% cost per watt reduction despite an increase in volumes sold from our higher-cost Ohio facilities and an increase in sales rate costs.
Thirdly, early generation First Solar CadTel modules that were installed at an NREL test facility in 1995 reached an installed life of 25 years and demonstrated a 25-year degradation rate of 48 basis points per year. While our manufacturing processes, product design, efficiency and warranted long-term degradation rates have improved significantly over the past 25 years, this result help us understand a legacy performance baseline and provides further confidence in the superior long-term durability and degradation performance of today's Series 6 product.
Fourthly, we extended our limited power output warranty from 25 to 30 years for our Series 6 modules and our Series 6 modules are now protected by the industry's first and only product warranty that specifically covers power loss from cell tracking, which can have a meaningful impact on reducing systems insurance costs.
Finally, as of year-end, we had shipments of 5.5 gigawatts, bookings of 5.5 gigawatts and contracted an additional 0.7 gigawatts of volume that remain subject to conditions precedent. Overall, our operational and financial results in 2020 have built momentum as we move into 2021.
Turning to Slide 4. I'll provide an update on our Series 6 capacity ramp and manufacturing performance. Over the course of 2020, we realized significant operational improvements. Comparing December fleet-wide metrics year-on-year, megawatts produced per day increased to 17.3 megawatts, an increase of 23%. Fleet-wide capacity utilization increased to 117%, an increase of 20 percentage points. Product yield increased to 97.6%, an increase of 3.2 percentage points. Average watts per module increased to 439 watts, an increase of 9 watts. And as noted, our top production bin increased to 445 watts. Our manufacturing discipline and execution enabled us to achieve our cost per watt reduction objective for the year.
We exited 2020 with 6.3 gigawatts of nameplate manufacturing capacity. And effective January 1, we have re-rated our throughput entitlement for purposes of calculating capacity utilization. Since launching Series 6 less than 3 years ago, the factory throughput entitlement was based on the initial tool set and factory design. Given the significant improvements made over the years, we have revised our throughput entitlement to reflect the 2020 exit rate throughput.
Our strong execution has continued into 2021 with improvement across all key metrics since year-end. In addition to February, we commenced additional production of our Series -- our second Series 6 low-cost factory in Malaysia. With less than 3 weeks of production, the factory is ramping nicely with demonstrated capacity utilization reaching approximately 80%, yield in excess of 90% and a top production bin of 450 watts. By the end of the year, we anticipate our Malaysia factories will have a nameplate capacity of 3 gigawatts.
Touching briefly on our systems segment. In February, we completed the sale of our 150-megawatt AC Sun Streams 2 project to Longroad Energy. We also signed agreements with Longroad to sell the Sun Streams 4 and 5 projects and are in late-stage negotiations to sign an agreement to sell our Sun Streams 3 project. As part of this portfolio acquisition, Longroad intends to utilize 1 gigawatt of Series 6, of which 785 megawatts will represent new bookings upon the closing of these transactions.
Prior to signing the potential agreement to sell Sun Streams 3, the project PPA was terminated, which enabled Longroad to include Sun Streams 3, 4 and 5 projects and their power marketing efforts after transactions close. While this resulted in approximately 85-megawatt systems de-booking in February, at the time of closing, we expect this opportunity will be re-recognized as a new module-only booking.
Turning to Slide 5. I'll next discuss our most recent bookings in greater detail. Our recent bookings momentum has continued with 3.3 gigawatts of net bookings since the October earnings call. After accounting for shipments of approximately 1.8 gigawatts during the fourth quarter, our future expected shipments, which extend into 2024, are 13.7 gigawatts. The majority of the bookings since the prior earnings call have been third-party module sales, which totaled 3.3 gigawatts.
We continue to see an increase in multiyear module sales agreements driven by our customers' need for certainty in terms of technology they are investing in and their suppliers' integrity and ethics. Representative of this, we have executed an agreement with Intersect Power to supply up to 2.4 gigawatts for deployment in projects in 2022 and 2023, of which approximately 2 gigawatts is recognized as a booking. In addition to this new booking, Intersect has the option to utilize an additional 0.4 gigawatts of module volume to support their portfolio project up to 2.4 gigawatts.
We've also secured 340 megawatts for deliveries in 2023 with a leading provider of hydrogen fuel cell solutions. A pillar of growth for the hydrogen economy is the ability to cost-effectively produce large-scale green hydrogen with renewable energy sources. With an environmentally advantaged CadTel technology, we are well positioned to address this market need.
Additionally, in Japan, we have continued success adding to our contracted systems backlog with the addition of 2 projects, totaling 51 megawatts.
With new net bookings of 3.3 gigawatts and with additional 1.4 gigawatts of expected bookings associated with the closing of the sale of the Sun Streams portfolio and the U.S. project development business, we are pleased with the robust demand for our Series 6 product. Included in these new bookings, volumes contracted to conditions precedent and the potential 0.4 gigawatts of incremental volume related for the Intersect transaction, we have 7.2 gigawatts of volume for potential deliveries in 2021, 5.9 gigawatts in 2022 and 2.3 gigawatts across 2023 and 2024.
Overall, while the market remains competitive, we are very pleased with the pricing levels that we are securing to date for our differentiated Series 6 Plus and CuRe modules.
In an industry that sells electrons and where products are evaluated based on the quantity of electrons they will produce, we also seek to differentiate our business model through our commitment to an environmental footprint of our technology, product secularity and supply chain transparency. We call it responsible solar, and you can learn more about it at our corporate website.
Turning to Slide 6. I'd like to discuss the strategy and advantages of this approach. Firstly, due to our resource-efficient manufacturing process, our thin film modules have the lowest carbon and water footprint available in the market today. With this advantaged position, Series 6 is the world's first PV product to be included in the EPEAT register for sustainable products, which conforms to the NSF 457, the industry's first sustainability leadership standard. Designed to help institutional purchasers, EPEAT is used by national governments, including the United States, and thousands of private sector institutional purchasers worldwide as part of their sustainable procurement decisions.
Secondly, we have over a decade of experience in operating high-value PV recycling facilities on a global scale and remain the only solar manufacturer to have global in-house recycling capabilities. This recycling process establishes a circular economy by recovering more than 90% of the semiconductor materials for reuse in First Solar's modules and 90% of the glass for use in new glass container products.
Thirdly, our vertically integrated manufacturing process enhances our supply chain transparency and control over our end-to-end manufacturing process.
We believe that our responsible solar strategy is the right way to do business and in a growing number of markets, yields an economic advantage. For example, France already has a rule that favors PV modules with a low carbon footprint. Spain also appears to be moving towards incorporating the carbon footprint metric in its renewable energy procurement program. A recent update requires owners of renewable energy generation assets to submit carbon footprint data to the country's renewable energy registry, gathering the information needed to shape the procurement mechanism that may benefit low carbon solar.
In the United States, Vectren, a utility that services Indiana and Ohio, included an environmental emission minimization objective within their integrated resource plan. This objective accounts for the cradle-to-grave emissions impacts of different forms of generation, including the low carbon footprint of thin film PV modules compared to crystalline silicon. Addition, Alliant Energy and Consumers Energy, 2 utilities in the Midwest, have included the aforementioned NSF 457 sustainability leadership standard for PV modules and inverters in their most recent solar solicitation.
We would also like to take the opportunity to touch on the reported use of forced labor in China's polysilicon manufacturing industry. We have repeatedly and unequivocally condemned the purported use of forced labor in China's PV solar supply chain and will continue to do so as long as it remains an issue. We also reiterated our commitment to zero tolerance of forced labor throughout our supply chain. We believe there should be no place for a solar panel where even a single component, no matter how small, is produced by a human being against their will.
We have seen reports that authorities in the United States are developing plans to expand their Xinjiang-specific import regulations to include solar. And in the latest version of the Forced Labor Prevention Act Bill, the U.S. House of Representatives included polysilicon as a high priority sector.
We recognize the challenges that this potentially creates for companies that have traditionally relied on Chinese-based firms for their modules, but as an industry, we cannot accept a view of solar at any cost. This is an important reminder that overreliance on China to supply subsidized solar panels comes at a price that may not always be reflected on the bottom line. It is a price that many -- may include needing to look the other way on environmental, social and human costs. It's also yet another reminder, one of several we've had this past year, about the importance of diversity of supply.
Before turning the call over to Alex, I would like to provide additional context on the effects of tariffs on the U.S. and global PV markets. In December 2012, during the Obama-Biden administration, the United States imposed antidumping and countervailing duties after determining that domestic crystalline silicon industry was materially injured by imports of crystalline silicon cells and modules that were sold at less than fair value and subsidized by the government of China. In March 2019, the United States continued these tariffs. There also was a second set of antidumping and countervailing duties on Chinese crystalline silicon modules with non-Chinese cells. Those duties were imposed in 2015 and in 2020 they were continued.
Given these tariffs only apply to a portion of crystalline silicon supply chain, Chinese manufacturers added cell and module capacity in nearby countries in Southeast Asia. Today, with this adjustment to their supply chain, our crystalline silicon competitors cannot only avoid these tariffs, but also continue to use government-subsidized polysilicon, ingots and wafers manufactured in China.
Separately, in February of 2018, during the Trump administration, the U.S. imposed Section 201 tariffs on imported crystalline silicon cells and modules from those countries with limited exceptions over a 4-year period. However, between June 2019 and November 2020, an exemption from Section 201 tariffs was granted for crystalline silicon bifacial modules. This exclusion enabled Chinese solar companies with bifacial cells and modules assembled in Southeast Asia to avoid the Section 201 tariffs as well as the antidumping and countervailing duties while they are still using subsidized polysilicon and ingots and wafers from China. Despite actions by the United States and India, most global markets have allowed unencumbered access of government-subsidized panels from China, resulting in PV economy and global goals that are largely beholden to a single technology supply chain and country.
We believe our differentiated technology and advantaged cost structure and a balanced perspective on growth, liquidity and profitability has enabled and will continue to enable us to succeed in the global marketplace despite the lack of fair trade. As the only alternative to crystalline silicon technology among the 10 largest solar module manufacturers globally, First Solar provides domestic supply security and enables the United States and global markets to reduce their overreliance on imported panels from China. We remain hopeful for a future where both free and fair trade can be established in the PV industry.
I'll now turn the call over to Alex, who will discuss our Q4 and full year 2020 results.
Alexander R. Bradley - CFO
Thanks, Mark. Starting on Slide 7, I'll cover the income statement highlights for the fourth quarter and full year 2020.
Net sales in the fourth quarter were $609 million, a decrease of $318 million compared to the prior quarter. This was primarily a result of higher international project sales in Q3, partially offset by increased module volumes sold in Q4. For the full year 2020, net sales were $2.7 billion compared to $3.1 billion in 2019. Versus our guidance expectations, net sales were within but towards the lower end of our guidance range.
This result was primarily caused by factors cited in our Q3 earnings call, which included the timing of the Sun Streams 2 project sale. To a lesser extent, net sales were also impacted by certain module deliveries that were delayed due to COVID-19-related events, including a positive case at a customer construction site which resulted in a temporary shutdown and a shipping vessel containing First Solar modules that was diverted from its intended destination due to a positive case on the vessel.
As a percentage of total quarterly sales, our module revenue in the fourth quarter was 90% compared to 46% in the third quarter. For the full year 2020, 64% of net sales were from our module business compared to 48% in 2019. Gross margin was 26% in the fourth quarter compared to 32% in the third quarter. And for the full year 2020, gross margin was 25% compared to 18% in 2019.
Systems segment revenue was $61 million in the fourth quarter compared to $505 million in the third quarter and fourth quarter systems revenue was lower than anticipated, primarily due to the delay in the sale of the Sun Streams 2 project. The systems segment gross margin was 18% in the fourth quarter compared to 33% in the third quarter. And the fourth quarter was positively impacted by a $9 million benefit associated with a reduction in estimated liquidated damages for legacy EPC projects, which increased systems segment gross margin by 14%. For the full year, systems segment gross margin was 26% compared to 16% in 2019.
The module segment gross margin was 27% in the fourth quarter compared to 30% in the third quarter. As a reminder, the third quarter was impacted by a reduction in our product warranty liability reserve, a reduction in our module collection and recycling liability and impairments of certain module manufacturing equipment for tools no longer compatible with our long-term technology roadmap. On a net basis, these factors increased Q3 module segment gross margin by 5 percentage points.
Also as a reminder, sales rate warranty are included in our cost of sales and reduced module segment gross margin by 7% in the fourth quarter compared to 6% in Q3. Despite utilizing contracted routes, minimizing changes and the use of the distribution center, we incurred higher rates during the fourth quarter for a portion of our module deliveries due to constrained container viability in the global shipping market. With this context in mind, we're pleased with our Q4 module segment gross margin results, which achieved our guidance expectation.
For the full year, module segment gross margin was 25% compared to 20% in 2019. Full year 2020 module segment gross margin included $20 million of severance and decommissioning costs and $4 million of ramp-related expense, which, in the aggregate, reduced module segment gross margin by 1.4%. From a fleet-wide perspective, as a result of our continued manufacturing execution, cost per watts sold at the end of 2020 met our target of a 10% decline relative to the end of 2019.
SG&A, R&D and production start-up totaled $102 million in the fourth quarter, an increase of approximately $16 million relative to the third quarter. This increase was primarily driven by an increase in production start-up expense from $13 million in Q3 to $17 million in Q4, $9 million of development project impairment charges in Q4 and a $7 million increase in incentive compensation expense relative to our guidance expectations, partially offset by the cost savings. With this context in mind, we're pleased with our operating expense result relative to our fourth quarter guidance range of $90 million to $95 million.
SG&A, R&D and start-up totaled $357 million in 2020 compared to $348 million in 2019. Included in the full year 2020 OpEx were $41 million of production start-up expense, $12 million of development project impairment charges, $7 million of severance charges, $6 million of class action and opt-out action legal fees, $3 million of expected credit losses on our accounts receivable as a result of the economic disruption caused by COVID-19 and $2 million of retention compensation expense. Combined with litigation losses of $6 million, total operating expenses were $363 million for full year 2020. Operating income was $58 million in Q4 and $317 million for the full year 2020.
We recorded a tax benefit of $66 million in the fourth quarter, which included a discrete tax benefit of $61 million associated with the closing of the statute of limitations on uncertain tax positions. For the full year, we recorded a tax benefit of approximately $107 million, which includes a full year net benefit from the CARES Act of approximately $84 million and $24 million related to the release of evaluation allowance in a foreign jurisdiction. During the fourth quarter, within equity and earnings, we recorded a full impairment of approximately $3 million related to one of our equity method investments.
Fourth quarter earnings per share was $1.08 compared to $1.45 in the prior quarter. For full year 2020, earnings per share was $3.73 compared to a loss per share of $1.09 in 2019.
Next on to Slide 8 to discuss select balance sheet items and summary cash flow information. Our cash and cash equivalents, restricted cash and marketable securities balance at year-end was $1.8 billion, an increase of $123 million from the prior quarter. Our net cash position, which includes cash and cash equivalents, restricted cash and multiple securities less debt, at year-end was $1.5 billion, an increase of $105 million from the prior quarter.
Our net cash balance is higher than our guidance due to lower-than-expected project spend on U.S. international development projects, the timing of cash payments for CapEx was delayed to the first quarter and improved collections on module sale agreements. Now due to the contemplated payment structure, the timing of the Sun Streams 2 project sale did not have a significant impact on our year-end cash balance relative to our guidance.
Cash flows from operations were $37 million in 2020 compared to $174 million in 2019. Cash flow from operations in 2020 include the previously disclosed payment of the class action opt-out litigation settlement of $369 million and a decrease in module prepayments following an increase in Q4 '19 associated with ITC safe harbor module purchase orders.
Also as a reminder, when we sell an asset at project-level debt that is assumed by the buyer, the operating cash flow associated with the sale is less than if the buyer had not assumed the debt. In 2020, buyers of our projects assumed $137 million in debt from these transactions.
Capital expenditures were $89 million in the fourth quarter compared to $106 million in the third quarter. Capital expenditures were $417 million in 2020 compared to $669 million in 2019.
And finally, before turning the call back over to Mark, I'd like to provide an update on the strategic review of our U.S. project development and North American O&M businesses. As recently announced, we signed a definitive agreement to sell our U.S. project development platform to Leeward Renewable Energy, a portfolio company of OMERS Infrastructure. This agreement follows a comprehensive, multi-phase process where more than 160 parties were either contacted or expressed inbound interest when multiple structures were considered. Based on the extensive nature of this process and the offers that we received, we believe this transaction represents the most compelling option.
We're pleased that the platform will be acquired by Leeward and almost the entirety of our U.S.-based project development team is expected to join Leeward upon closing. The transaction is expected to close in the first half of 2021 after obtaining regulatory approval and satisfying customer and closing conditions.
Subject to closing the acquisition, Leeward will sign or assume 1.8 gigawatts of module purchase orders, of which 744 megawatts represent new bookings. While approximately 0.4 gigawatts are included in the upfront purchase price, the remaining approximately 1.4 gigawatts of modules are expected to be added to our contracted backlog and are expected to be recognized as future module segment revenue.
As previously noted in our U.S. project development sale announcement, we stated that we intended to retain 1.1 gigawatt AC of U.S.-based projects that we plan to sell separately. After this announcement, we closed the sale of our Sun Streams 2 project, signed agreements to sell our Sun Streams 4 and 5 projects and are in late-stage negotiations to sell our Sun Streams 3 project, which totaled 750 megawatts AC. The remaining projects are uncontracted and are expected to be sold in 2021.
As it relates to the sale of our North American O&M business to NovaSource Power Services, a portfolio company of Clairvest Group, although we initially expected the sale of this business to close in the fourth quarter of 2020, certain conditions to closing remain outstanding. We expect these remaining conditions to be satisfied and the transaction to close in the first half of '21.
I'll later discuss the financial impacts of these transactions during the guidance portion of today's call. Now I'll turn it back over to Mark to provide the business and strategy update.
Mark R. Widmar - CEO & Director
All right. Thank you, Alex. With our company's founding over 20 years ago, the PV industry has been through periods of rapid growth, declining costs and technology evolution. We're one of the few solar companies that both entered and exited this last decade. We have continued to adapt our business model to remain competitive and differentiated in a constantly evolving market. For example, our original insets into O&M and EPC and project development was to address an unmet need of the market and capture a profit pool. Our acceleration of Series 6 production was a competitive response to address the current market condition. Despite these transformation, among others, our core identity as a module manufacturing company with a differentiated CadTel technology has remained constant.
As we've looked into the future with a more focused business model, our pace of innovation will be critical to our competitive strengths, enabling us to leverage our points of differentiation and capture compelling value for our technology. Sure, with an attractive warranty and responsible solar strategy are recent examples of innovations enhancing our competitive position in the market.
This market -- the market momentum for PV continues to build. Our Series 6 energy, quality and environmental advantages are all key differentiators, which we believe will enable us to meaningfully participate in this wave of demand that's seen in renewable energy. Based on the growth of selected PV markets and our competitive advantages, we believe we can grow our manufacturing capacity while still selling our products into regions where our technology has points of differentiation. Within this context, Slide 9 provides an updated view of our global potential bookings opportunity, which now totals 19.7 gigawatts across early to late-stage opportunities through 2023. In terms of segment mix, this pipeline of opportunity is exclusively third-party module sales.
In terms of geographical breakdown, North America remains the region with the largest number of opportunities at 14.9 gigawatts. Europe represents 2.3 gigawatts, India represents 1.8 gigawatts with the remainder in other geographies. A subset of this opportunity set is our mid- to late-stage booking opportunities of 12.6 gigawatts, which reflects those opportunities we feel could book within the next 12 months and include the aforementioned 1.4 gigawatts of contracted subject to satisfaction of commission's president. This subset includes approximately 10.2 gigawatts in North America, 1.2 gigawatts in India, 0.9 gigawatts in Europe, of which 0.7 gigawatts is based in France and the remainder in other geographies. This opportunity set, coupled with our contracted backlog, gives us confidence as we continue scaling our manufacturing capacity.
Turning to Slide 10, as we continue to drive additional throughput, increased average watts per module and improved manufacturing yield, our Series 6 production exited 2020 with nameplate capacity manufacturing of approximately 6.3 gigawatts, split between 4.1 gigawatts at our international factories in Vietnam and Malaysia, and 2.2 gigawatts in Ohio. With the commenced production at our second Series 6 factory in Malaysia, our global manufacturing footprint increases to 6 factories. At the end of 2021, we anticipate increasing nameplate capacity to 8.7 gigawatts, which includes 2.6 gigawatts of capacity in Ohio, and 6.1 gigawatts across 4 factories in Malaysia and Vietnam. This 2.4 gigawatts of incremental year-over-year capacity is reflective of our new Malaysia factory and expected improvements in average loss per module and throughput across the fleet. By the end of 2022, we anticipate increasing throughput by 12% compared to our re-rated throughput entitlement and expect continued improvements in our average loss per module and manufacturing yield. Accordingly, by the end of the year, we anticipate increasing our fleet-wide nameplate manufacturing capacity to 9.4 gigawatts, which includes 2.6 -- 2.7 gigawatts of capacity in Ohio and 6.7 gigawatts across our international factories. This 0.7 gigawatts of anticipated incremental capacity is expected to come from optimization of our existing footprint. As previously highlighted, we are evaluating the potential for future capacity expansion and they seek to further diversify our manufacturing presence. In addition to the factors we've previously highlighted, we are also evaluating domestic and international policies to ensure any such expansion is well positioned. While we have made no such decisions at this time, any greenfield capacity addition are unlikely to contribute to our 2022 production plan.
From a production perspective, in 2021, we expect to produce approximately 7.4 to 7.6 gigawatts, which is within the 7.3 to 7.7 gigawatt range we have provided the last -- at this time of the last February guidance call. Note, our second Malaysia factory will continue its ramp period through the end of the first quarter, and we are planning for over 3 weeks of downtime across the fleet to implement technology and throughput upgrades.
In 2022, with the addition of the fully ramped factory in Malaysia and ongoing improvements across the fleet, we expect to produce 8.6 to 9.0 gigawatts.
Turning to Slide 11. I will now provide an update on our technology road map. Over the course of 2020, we have made steady progress in our technology road map, ending the year with a top bin of 445. Early in 2021, we have demonstrated continued progress increasing our fleet-wide average per module to 440 for February month-to-date. And for our new Malaysia factory, introduced our Series 6 Plus module, the next phase of our technology road map with the current top bin of 450 watts.
Leveraging our existing Series 6 tool set, we increased our module form factor by approximately 2% and increased our module efficiency, which has increased our top bin production by approximately 10 watts. Note, after our second Malaysia factory ramp is completed, we anticipate our top bin will be 455 watts. Importantly, this increase in form factor is sized to reduce balance of system cost per watt by adding moderate wattage without material changes to the installation process or support structure. We anticipate implementing Series 6 Plus across the city over the course of 2021. From a manufacturing cost perspective, we expect this additional wattage will reduce our cost and sales freight per watt, which I will later discuss.
For the fourth quarter of 2021, we anticipate commencing initial production of our copper-replaced Series 6 or CuRe on our lead line production. As previously disclosed, this program is expected to not only increase module wattage, but also meaningfully improved lifetime energy performance. Accordingly, by the end of 2021, we anticipate our top production bin will reach $460 million to $465 million. With an expected 30-year warranty degradation rate approximately 50% below our existing baseline.
Given TV power plants have an expected useful life of up to 40 years, a reduction in a module's long-term degradation is expected to be a material benefit to project economics as it increases energy density of the module and life cycle energy generation.
As demonstrated on Slide 12, we believe the benefits of improved module efficiency and temperature coefficient will result in a 7% higher energy density in the first year for our 465-watt CuRe module compared to our 440-watt Series 6 module. Due to the expected reduction in our CuRe modules long-term degradation rate, we expect this improvement can increase to 20% in year 40, which represents a 13% improvement over the life of the asset. As we've stated previously, we believe CuRe significantly increases Series 6's competitiveness against bifacial modules. As a point of reference, bifacial modules generated an estimate 4% to 8% more energy than comparable monofacial modules. More importantly, CuRe's energy uplift does not increase the module or balance of system cost is typically seen with bifacial modules.
By the end of the first quarter of 2022, we anticipate the entire fleet will be converted to CuRe. This is anticipated to provide additional benefits to our average watt per module and cost per watt.
Through the implementation of our copper replacement program, combined with our ongoing R&D program, we're aiming to achieve a top production bin of 475 to 480 watts by the end of 2022. So on our second quarter earnings call, we stated that we expected a 480-watt module bin in 2023. With a CadTel cell efficiency entitlement in excess of 25%, we see a path to significantly increase our module wattage and efficiency in the midterm. With this path to increased efficiency, coupled with our degradation, spectral response and temperature coefficient energy advantages and vertically integrated manufacturing processes, we believe the outlook for our technology remains well positioned in a global PV market.
Finally, we continue to focus on advanced research and development and are evaluating the potential to move beyond a single-junction device and leverage the high band gap advantages of CadTel in a multi-junction device. A multi-junction device has the potential to be a disruptive, high efficiency, low-cost module with an advantaged energy generation profile. While the evaluation of this technology is in early development, we are aiming to utilize many of the product enhancements in our existing CadTel road map.
Turning to Slide 13. I'll provide some context around our module cost per watt. As initially presented on our guidance call in February 2020, we forecasted a Series 6 cost per watt reduction of 10% between where we expected to end 2020 and the end of 2019. Despite unforeseen challenges related to the pandemic, pricing pressures in the global shipping market and rising commodity costs including aluminum, which we mitigated in part through a hedge structure, and increased demand for PV glass, we executed on our cost per watt road map for the year and achieved this target.
Looking into 2021, I'd like to start by addressing how we intend to manage key bill of material and sales freight costs. Firstly, given our module utilizes a CadTel chemistry, our cost per watt is unaffected by fluctuations in polysilicon pricing. Secondly, from a glass perspective, growing solar demand and the emergence of bifacial modules have continued to put pressure on the supply and cost of PV glass. However, our glass procurement strategy primarily relies on forward contracts and localization of glass supply. In 2021, we intend to further localize our glass needs domestically in the United States and Malaysia through long-term supply agreements. This strategy enables us to mitigate the cost of variable spot pricing for glass and inbound freight.
Thirdly, from a sales freight perspective, utilizing contracted routes and minimizing changes helped alleviate some of the impact of higher spot rates in 2020 in the first quarter of 2021. Despite higher shipping rates expected in 2021, we intend to utilize our distribution center strategy to mitigate some of these impact. Note, we expect sales freight and warranty to reduce module segment gross margin by 7 to 8 percentage points in 2021 compared to 7 percentage points in 2020.
Finally, as part of our Series 6 Plus implementation, we anticipate a reduction in the module profile by reducing the thickness of our frame injunction box. In addition to reducing the bill of material costs, we anticipate this development will enable us to increase module shipping container -- modules per shipping container by approximately 10%.
As it relates to our Ohio manufacturing facilities, despite exiting 2020 with a higher cost per watt in comparison to our international factories, we anticipate significant improvements in 2021 through the following initiatives: Firstly, in the fourth quarter, the manufacturing yield was 96%, which was below the fleet average. We anticipate this will improve to 97% by the end of 2021, which provides a benefit to our fixed and variable cost per module; secondly, we anticipate increasing our nameplate manufacturing capacity to 2.6 gigawatts by the end of the year, an increase of 18% compared to the end of 2020; finally, a covered glass facility in Illinois started in the fourth quarter of 2020 and our float glass facility in Ohio started in the first quarter of 2021 and will supply our Ohio factory. We anticipate this will provide a benefit to the variable portion of our cost per watt. Through the implementation of these key initiatives, among others, we anticipate our Ohio cost per watt headwind relative to our international factories will exit 2021 $0.02 per watt higher, including sales freight.
On a fleet-wide basis relative to where we exited 2020, we anticipate reducing our cost per watt produced by 11% by the end of 2021. Due to the ramp in other underutilization costs related to the aforementioned factory ramp upgrades and challenges related to sales freight, we anticipate reducing our cost per watt sold by 8% by the end of the year.
As we look beyond the midterm, I would like to revisit the 5 key levers that we believe will enable us to continue reducing our cost per watt. Starting with efficiency, we anticipate increasing our top production bin from 445 in December 2020 to a top production bin of 475 to 480 watts by the end of 2022. With a midterm goal of 500 watts per module, we see the potential for continued improvement in our module performance. Improvements in module watts generally provide a benefit to each component of the cost per watt, including our variable and fixed bill of materials and sales freight and warranty costs.
Secondly, by the end of 2022, we anticipate increasing throughput by 12% compared to our re-rated capacity utilization baseline through the implementation of additional tools and debottlenecking efforts. This drives additional throughput on our existing manufacturing footprint resulting in the fixed cost dilution benefit. Thirdly, while we've made steady improvements to our manufacturing yield over the course of 2020, achieving 97.6% in December, we anticipate a fleet-wide yield of 97.5% in 2021.
While our international factories have achieved yield in excess of 98%, the planned upgrades for Series 6 Plus and CuRe are expected to impact yield performance during the year. However, in the midterm, we see a path to increase our fleet-wide manufacturing yield to 98.5%.
Fourthly, we see midterm opportunities to reduce our bill of material costs by 20% to 25%, primarily across our glass and frame. Finally, we believe the combination of fitting our model profile and transportation optimization can lead to a 15% reduction in freight costs. Combining the benefits of our CuRe and our other R&D work with aforementioned cost levers, we believe we are strongly positioned to continue to drive Series 6 cost per watt efficiency and energy improvements over the near and midterm.
I'll now turn the call back over to Alex, who will discuss our financial outlook and provide 2021 guidance.
Alexander R. Bradley - CFO
Thanks, Mark. Before discussing our 2021 financial guidance, I'd like to highlight our core operating [pens], which is endeavored to create shareholder value through a disciplined decision-making framework, the balances growth, liquidity and profitability. As it relates to growth, we anticipate increasing our nameplate manufacturing capacity to 9.4 gigawatts by the end of 2020, driven by the addition of our second factory in Malaysia and ongoing improvements in average watts per module throughput manufacturing yield. As Mark previously highlighted, we're evaluating the potential for future capacity expansion, and may do some beyond our existing geographic footprint.
Strong bookings performance in 2020 and year-to-date 2021 and current forward contract position of 13.7 gigawatts gives us commercial confidence as we evaluate the potential for incremental expansion.
Our liquidity position has been a strategic differentiator in an industry that has historically prioritized growth without regard to long-term capital structure. For example, one of the few solar companies that both entered and exited the last decade and our strong balance sheet has enabled us to weather periods of volatility and also pursue growth opportunities. Additionally, we were able to self-fund our Series 6 transition whilst maintaining our strong liquidity position, ending 2020 with $1.5 billion of net cash. And to say, we'll be able to continue to self-fund future capacity expansion and strategic investments from our technology whilst maintaining strong differentiated balance sheet, which we believe is a meaningful competitive differentiator.
And from a profitability perspective, our contracted backlog provides increased visibility into future sales, reduces financial exposure to spot pricing for PV modules, helps align our capacity with future demand. Accordingly, we can be selective with our bookings opportunities and contract module sales at pricing levels that fairly value our Energy Advantage product and provide an acceptable profit per watt. For example, in 2022, although there remains significant uncontracted volume yet to book, the ASP across the aforementioned 5.9 gigawatts of volume for potential deliveries in 2022 is only 10% lower than that of the 7.2 gigawatts of volume to be shipped in 2021. With a target of 11% reduction in costs per watt produced between year-end '20 and year-end '21, we believe there's an opportunity to capture an attractive margin.
So with this context in mind, I'll next discuss the assumptions included in our 2021 financial events. Please turn to Slide 14. As it relates to our U.S. project development business, we anticipate that the transaction will close in the first half of 2021, the expected proceeds of approximately $270 million. Included in this price of 390 million -- sorry, 390 megawatts of Series 4 and Series 6 Plus solar modules, the 10-gigawatt project pipeline, including the 5 contracted development projects, the 30-megawatt operational Beryl project and certain other safe harbor equipment.
On closing, we expect to recognize a pretax gain on sale shown on the income statement between the gross margin and operating income lines of approximately $25 million. As it relates to our North American O&M business, we anticipate this transaction will also close in the first half of 2021. And upon closing, we expect to recognize a pretax gain on sale of approximately $115 million. We believe the closing of these transactions will be a positive result for both our U.S. project development and North American O&M associates.
As of the end of 2020, we had approximately 300 associates that supported our North American O&M and U.S. portion businesses collectively. And at closing, substantially all of these associates will join Leeward and NovaSource, respectively.
As we exit North American O&M and U.S. project development, we see the potential for significant cost reductions from these decisions, which is reflected in both the cost of sales and operating expenses lines. As we mentioned on prior earnings calls, including Q3 of 2019 and as is also the case in Q4 of 2020, in quarters with low project development revenue, we see an adverse impact to Systems segment gross margin due to the fixed cost burden that sits in the cost of sales line. Similarly, for the O&M business, the majority of the non-direct project-related costs to support the O&M business sit within the cost of sales line.
In total in 2021, we expect to see approximately $15 million in annual cost of sales savings associated with the sale of the North American O&M business. With an additional approximately $5 million of savings in 2022, we expect run rate annual savings of approximately $20 million from the sale from 2022 in North America.
The sale of the U.S. project development business is expected to result in approximately $35 million of savings in 2021, and an additional $10 million to $15 million of run rate savings in '22, for a total annualized benefit from '22 onwards approximately $45 million to $50 million, of which approximately 60% sits in the operating expenses line.
From a systems perspective, remaining in our 2021 cost structure are approximately $15 million of expenses associated with our Japanese development business, split between operating expense and cost of sales, and approximately $15 million of cost of sales associated with our power generating assets.
For the booked backlog, system backlog of approximately 200 megawatts AC of systems projects in Japan and a strong capacity position will leave us an opportunity to capture an attractive profit pool there.
Next, our 2021 shipments are expected to be between 7.8 and 8 gigawatts, which exceeds our production plan for the year, of 7.4 to 7.6 gigawatts. There are several factors driving this. Firstly, we produced approximately 1.6 gigawatts in the fourth quarter, which exceeded our guidance from the third quarter earnings call by about 120 megawatts.
Secondly, in the fourth quarter, we shipped 1.8 gigawatts, which was 100 megawatts below the midpoint of our guidance range. And finally, we expect to ship approximately 150 megawatts of Series 6 modules as part of the U.S. product development transaction that were previously intended to safe harbor the 26% investment tax benefit.
Our ongoing Series 6 throughput and technology programs are expected to impact 2020 an operating income by $60 million to $70 million. This is comprised of $5 million to $10 million of ramp expenses incurred at our second factory in Malaysia, which we anticipate will exit as ramp period by the end of Q1. As previously mentioned, we have fleet-wide factory upgrades to incorporate Series 6 Plus CuRe and throughput improvements in 2021. The upgrades will require approximately 3 weeks of downtime across the fleet, resulting in estimated underutilization losses of $40 million, and production start-up expense of $15 million to $20 million. We anticipate these improvements will contribute meaningfully to our 8.6 to 9 gigawatt production plan in 2022.
As it relates to domestic capital markets and financing, with significant utility scale, solar and wind capacity additions expected in 2021, with co-located battery storage increasing many projects ITC-eligible basis, demand for tax equity is at this time expected to remain high. Our financial guidance assumes that bank profitability will be sufficient to supply the needs of the tax equity market or if market conditions deteriorate, an appropriate legislative solution such as the ability to receive direct cash payments in lieu of investment tax credits is implemented.
And finally, to date, we have largely managed the impact of the COVID-19 outbreak on our business, and it has not had significant impacts on our operations. Our guidance accordingly assumes we will continue to be able to mitigate any such impact on our supply chain operations without the incurrent material costs.
I'll now cover the 2021 guidance ranges on Slide 15. Our net sales guidance is between $2.85 billion and $3 billion, which includes $2.45 billion to $2.55 billion of module segment revenue. Included in our systems revenue guidance is the sale of the Sun Streams 2 project, which closed in Q1.
Gross margin is expected to be between $710 million and $775 million, which includes $580 million to $625 million of module segment gross margin. Module segment gross margin includes a combined $45 million to $50 million of ramp expense and underutilization losses, which are expected to reduce module segment gross margin by approximately 2 percentage points. Additionally, sales freight and warranty are included in cost of sales and are expected to reduce module segment gross margin by 7 to 8 percentage points.
In the United States, we're seeing some weather-related impacts to module delivery schedules resulting from last week's storm, particularly in Texas. Whilst we're in the process of balancing customers' project needs and contractual commitments, we anticipate this will impact our Q1 shipments. However, with lower Q1 sales volume and an improving cost block profile over the course of the year, we anticipate our module 7 gross margin will increase from 19% in the first quarter to 26% in the fourth quarter. Approximately 1/3 of our full year ramp and underutilization charges are expected to be incurred during the first quarter. With the remainder split evenly across the subsequent 3 quarters.
To broaden out our sellable volume in 2021 is predominantly Series 6 and Series 6 Plus, which is competing for business against bifacial technology. Whilst we are incurring ramp in utilization costs this year to integrate our CuRe technology, we expect to begin realizing the value associated with these improvements in 2022.
SG&A and R&D expenses are expected to total $270 million to $280 million. Included in SG&A, approximately $5 million of transaction costs related to the sale of our U.S. project development business.
Operating expenses, which includes $15 million to $20 million of production start-up expense, are expected to be between $285 million and $300 million.
Operating income is estimated to be between $545 million and $640 million and is inclusive of an expected approximately $140 million gain on sale related to the aforementioned O&M and project development transactions. And $60 million to $70 million of combined ramp and underutilization costs and plant start-up expenses.
Certain nonoperating items, we expect interest income, interest expense and other income to net a negative $10 million. Full year tax expense is forecast to be $100 million to $120 million, which includes approximately $35 million of tax expense related to the North American O&M and U.S. project development sales transactions.
This results in a full year 2021 earnings per share guidance range of $4.05 to $4.75. And now we expect earnings per share of approximately $1 related to the gains on sale of our U.S. project development and North American O&M businesses.
Capital expenditures in 2021 are expected to range from $425 million to $475 million, as we complete the transition to our second Series 6 factory in Malaysia, increased throughput on our existing Series 6 facilities, implement Series 6 Plus and CuRe, and invest in other R&D related programs.
Our year-end 2021 net cash balance is anticipated to be between $1.8 billion and $1.9 billion. The increase from our 2020 year-end net cash balance is primarily due to operating cash flows from our modules business, the proceeds from our U.S. project development and North American O&M sales, which we expect will be partially offset by capital expenditures.
Turning to Slide 16, I'll summarize the key messages from today's call. We continue to make significant progress on our Series 6 transition, both from a demand and supply perspective. Series 6 demand was in robust with 3.3 gigawatts of net bookings in previous earnings call, an additional 1.4 gigawatt of volume contracted subject to conditions precedent. Our opportunity pipeline continues to grow with a global opportunity set of 19.7 gigawatts, including mid- to late-stage opportunities of 12.6 gigawatts.
On the supply side, we continue to expand our manufacturing capacity and expect to increase our nameplate Series 6 manufacturing capacity to 8.7 gigawatts by year-end 2021 and 9.4 gigawatts by year-end 2022. In 2021, we expect to produce 7.4 to 7.6 gigawatts of Series 6 volume, a year-over-year increase of 25% to 29%. And we see significant midterm opportunity for improvements to our module efficiency, cost and energy metrics. We ended 2020 with full year EPS of $3.73 and forecasting full year 2021 earnings per share of $4.05 to $4.75.
And finally, we expect to close the sale of our North American O&M and U.S. project development businesses in the first half of 2021.
And with that, we conclude our prepared remarks, and I'll open the call for questions. Operator?
Operator
(Operator Instructions) Our first question comes from Philip Shen with ROTH.
Philip Shen - MD & Senior Research Analyst
You've shown some healthy bookings year-to-date. Given the forced labor issue ramping up, can you talk about how recent conversations with customers have been shaping up and perhaps how they've shifted as well?
And then looking out to '22, when do you expect that could become fully booked? Looks like you're 2/3 there. And then what about the outlook for '23?
And then in terms of your recent bookings, you talked about, I think, a 10% reduction in pricing from 2020 levels, which might suggest that your 2022 bookings that you've gained or booked recently are in the $0.30 per watt. So I was wondering if you could comment on that? Or if they might be closer to the mid-20s cents per watt, which is, I think, certainly impossible on what some market participants have been sharing with us in terms of market pricing for crystalline silicon? So I know there's a lot there.
Mark R. Widmar - CEO & Director
All right, Phil. So I'll hit on all 3 of them. Look, on bookings, we're really happy with the momentum. And just if you even look at the mid- to late-stage of opportunities for -- we expect opportunities which we could close within the next year with north of 12 gigawatts sitting there. The momentum we're starting off with right now, we expect 2022 to be a very strong booking year. As it relates to this discussion and the comment around the implications of forced labor, I think we try to hit on some of those themes of -- and it's probably not just any 1 particular issue, but it goes back to this concept that we refer to as responsible solar. And there clearly are a number of counterparties and customers that we have engaged with in conversations that are, one, concern about overreliance, concern about maybe the current state of political relationships between the U.S. and China or India and China or other markets as well. And as a result of that, they're looking for alternatives. And one thing that's great about First Solar, not only do we have great technology and great capabilities, but having a different standard, which we hold ourselves accountable for. And we have different value attributes that we can provide to our customers and certainty is one, in dealing with a counterparty or a supplier with a different -- it held themselves from an integrity and ethical standard to the highest levels. I think it's important. And it's starting to come into the bookings and what we've started to see now in the pipeline that we have. So Phil, it's one of many. I think there's still a lot of people that are trying to understand the whole forced labor and how it plays out and what the potential applications are around it.
But what I will tell you is that some of end customers that we have, not referring to here as the IPP or developer or EPC, but others that are more of the off-take agreements. They're very concerned, and in some cases, they're incorporating conditions within their procurement criteria to ensure that there's 0 tolerance for forced labor. And not only the modules of which are being procured and utilized in the project that they have in the U.S. or somewhere else internationally, they want to make sure that their suppliers also have 0 tolerance, and there's nowhere through their entire supply chain do they tolerate forced labor. And that makes it very hard as you know with the complexity of crystallin supply chain to make those types of assertions and representation.
As it relates to our goals, and I'll be very transparent on our goals. I'm not going to give you a discrete timing, but we clearly want to be -- as we exit this year, our goal is we want to be sold out of '22. We want to be more than halfway, sold out at '23, and have a meaningful portion of our '24 volumes contract.
So -- and when you look at where we are right now, and as you've indicated, we're pretty -- in pretty good position for '22. We have more work to do. And I keep telling our Chief Commercial Officer, let's keep selling. Keep selling and taking the opportunities that we can. And it's good to see the pipeline and robustness and resiliency that we have that hopefully gets us to that goal of accomplishing that by the time we exit the year.
And that clearly would put us at a much higher than we historically have tried to say, let's maintain a 1:1 sold to book ratio. If we accomplished that, we'll be much higher than our historical 1:1. And at least where we sit right now, we feel confident we can get that. ASPs, what I'll say, Phil, around that, is that -- I mean, if you look at the Q when it comes out or the K in this case, I think it's going to come out with an average of like $0.308 or something like that, the metrics, I'm going to tell you that. The -- I wouldn't say that we're starting to see increase in ASPS. I wouldn't say that's necessarily a fact of what we're seeing. Are we starting to see -- it depends on each year, which we're booking out into, ASP is starting to firm up. And maybe we'll start to see some resiliency in the upward direction. But like I said, we said on the call, is we're very pleased and happy with the ASPs that we've been able to capture relative to the value we create in our technology. And then the opportunity to continue to improve our overall cost reduction road map, maintain a solid gross margin. But more importantly, as you look at year-over-year, we're growing our capacity and our sold volumes will be up significantly year-on-year as well with all sort of creates contribution margin that helps expand operating income.
Operator
Our next question comes from Brian Lee with Goldman Sachs.
Brian K. Lee - VP & Senior Clean Energy Analyst
Had 2 here. I guess, first, Mark, can you -- of the 1.1 gigawatts of systems, I think you talked about this, but how much is targeted to be sold this year, next year? And then presumably, 2023 will be the last year where you see some systems business revenue, and how much in that year? And then the gross margins, I know they depend a lot on mix, but it seems like if we back out the components where you're doing pretty well, it's about a high single digit, low teens number implied for this year. Is that going to be kind of the go-forward margin level? I would have thought it'd be a little bit higher given Sun Streams made it into '21 versus 2020, but any thoughts around mix, implications for margins and how to think about margins for that business as you still have some revenue to monetize over the next couple of years?
Mark R. Widmar - CEO & Director
Yes. I'll take the first one, Brian, and I'll let Alex take the question on gross margin. As it relates to the Systems business, the $1.1 billion is largely the U.S. assets that we still have. Sun Streams is contracted. The rest of the Sun Streams is complex we've signed, but we haven't finalized, what portion beside another portion hasn't signed yet, but the plan would have all that done, hopefully, by the end of the quarter. And that will take the largest portion of the U.S. volumes at the aggregates up to on an AC basis, I don't know, 600 megawatts or something like that. And there's another few hundred megawatts that's left in the U.S., which our plan would be to move forward as quickly as possible, and ideally have all of that sold out by the end of this year. So those are just development sites, which we then try to contract module. I'll take agreements, too. But the goal for the U.S. stuff would be to monetize all that volume this year, and sooner is the better. The development team is going with the transaction. So we don't have the capabilities really to continue with the development activities. We'll have to enter into a service agreement effectively for -- with Leeward to continue to support those projects until they're sold down.
The balance of it, there is still a 200 to 300 megawatts of contracted Japan projects. And there's still more that's not contracted at this point in time, we have feed-in tariffs that we haven't actually fully accomplished the permitting process and interconnection and other things that would ultimately -- we ultimately require for recognizing of the booking. That volume will be recognized. Most of it will show up in '23, there will be some in '21 and '22. But if you look at the CODs on those projects, most of them have '23 CODs as we currently see them. But as you know, the bookings in -- excuse me, the average ASPs on those projects are highly attractive. And so we'll monetize that over the next 3 years. And we'll see if we go beyond that. Again, we still have some more projects with fees and tariffs that we haven't booked yet that potentially could create more volume into the '23, maybe even '24 period. But I'll let Alex talk about the gross margin.
Alexander R. Bradley - CFO
Yes. So Brian, if you look at the guidance we gave, total revenue, $2.85 billion to $3 billion, of that module $2.45 billion to $2.55 billion, that implies systems of $400 million to $450 million of revenue. And on the gross margin side, on the $710 million to $785 million company-wide, module of $580 million to $625 million, so again implies systems $130 million to $160 million. So if you look at those gross margins, it's 25% to 26% at the company level. Systems looks pretty high, but it's really very limited volume there. So Systems is in the low 30s, skewed a little bit by Sun Streams and then potentially a little bit of Japan coming in back end of the year. But the module there comes in at about 24% to 25%. So that's where you're seeing module gross margin for the year.
And then as you're talking about kind of how to look at that going out, we tried to give a little bit of color here on the gross margin level, we talked about the ASP decline and what we've got booked, right? And if you look at '22, there's clearly still a lot to book in there. But we do have a significant amount contracted already. And we said if you look at the ASP decline, '21 and '22 goes down about 10% and then cost decline going down 11%. Now those are off different bases, obviously, on an ASP and the cost per watt. But you can see that we're getting cost per watt coming down at or slightly better than ASP decline going from '21 into '22. And at the same time, you're getting some additional volume coming through where so you get that scale.
On a percentage basis, you get a better benefit there from dilution of some fixed cost. But on an absolute dollar basis, you get just the benefit of increased volume coming through there.
And then as we talked about before, it matters also to go down to the operating margin level. So we talked about some of the cost reductions coming out from the sale of the O&M and project development business. Some of that comes out in 2021, but there's also a lack of a little bit that comes out in 2022. So that's going to see cost of sales and OpEx continuing to reduce as we go into '22, and that, again, helps us down to an operating margin level.
Mark R. Widmar - CEO & Director
Yes, the only thing I'll add to that, too, Brian. And we state Alex mentioned in his comments, it's just that there's a pretty significant headwind, at least in '21 for underutilization in order to deal with the upgrades that we need to do for -- primarily for CuRe. So there's a significant cost, I think it's about $40 million in total of underutilization that we'll have to absorb within '21. So that's weighing down on the gross margin. I think if you could adjust for that, I think the gross margin goes up a couple of percentage points up into that range.
Operator
Our final question will come from Ben Kallo with Baird.
Benjamin Joseph Kallo - Senior Research Analyst
I kind of want to follow it all down. So I heard you say, Mark, several different things about gross margin improving. You said the ASPs are firming up. You're much locked into '22. So if I go to '22, EPS should be up, right?
And then my second question is -- I guess, Alex, when you build a new factory, how do you determine whether or not you have the ROIC on that? And I guess there's probably a margin associated with that. And so you have to have some kind of firm belief in your long-term contracts to invest that money. And so those are my 2 questions.
Mark R. Widmar - CEO & Director
Yes, Ben, I'll take the first one, I'll let Alex take the second one. Ben, again, we're not giving guidance for '22 at this point. Now we gave some pretty strong indicators of what will drive '22, which will be the volume of production volumes, as we referenced, a new product of CuRe. The one thing we want to keep making sure that's represented in there is -- and all the '22 volume will be CuRe. If you look at the 1 slide, which shows the energy uplift, there's a meaningful energy uplift because of the improvement of long-term degradation. And that -- we sell energy. We sell value. And so that is important to understand. We also referenced that. Look, there was a lot of interest and rightfully so when bifacial modules came out and they talked about a 4% to 8% energy uplift, relative to monofacial modules of similar efficiency.
If you look at where we are with our CuRe product and on life cycle average, on top of the initial improvement of TEPCO and efficiency pop, there's another 10 percentage points of life cycle energy through improvement of long-term degradation. And so you can take our product. And even against crystalline silicon bifacial that maybe even has a nameplate of 150, 175 bps better efficiency. And you're going to find that over the life cycle energy profile, we're going to outperform that in the range of, call it, anywhere from 4% to 6%. And that's compelling value creation. So it's the technology, it's the supply improvement. There was a production plan improvement that we're talking about and continued reduction of our own internal costs. I think '22, we're not going to give specific guidance, but we gave enough information, I think, to help people look across the horizon and what '22 should look like.
Alexander R. Bradley - CFO
Yes. And Ben, I think about ROIC. So I look at it across both an individual factory and on a company-wide basis. And if you think about individual factory, at a gross margin level, depending on where that volume is sold, if more of it is sold outside the U.S. today and as we expand, you may see that gross margin level going down and being more challenging from an individual factory relative to the current book volume. But at the same time, adding a factory adds very limited to no OpEx, it also actually has a slight benefit in diluting some of the fixed costs, instead of the cost of sales line. So therefore, you -- on an operating margin benefit if anything we actually can look better. So because it has impacts not just for an individual factory but also for diluting fixed cost, cost per watt plus the benefit you get of extra volume was pretty much the same OpEx given that, as we said before, we think 80% to 90% of our operating cost line is fixed. We have to look at it both individually and across the company. But we certainly want to make sure that if we're adding, we're being significantly on our way to cost of capital for adding factories, and we've seen that today.
Operator
We have reached the end of our time for the question-and-answer session. This concludes today's conference call. You may now disconnect.