Stephen Haymore - First Solar, Inc. Mark R. Widmar - First Solar, Inc. Alexander R. Bradley - First Solar, Inc..

Paul Coster - JPMorgan Securities LLC Brian Lee - Goldman Sachs & Co. Tyler Charles Frank - Robert W. Baird & Co., Inc. (Private Wealth Management) Krish Sankar - Bank of America Merrill Lynch Vishal Shah - Deutsche Bank Securities, Inc. Colin Rusch - Oppenheimer & Co., Inc..

Good afternoon, everyone, and welcome to First Solar's Fourth Quarter 2016 Earnings Call. This call is being webcast live on the Investors section of First Solar's website at firstsolar.com. At this time, all participants are in listen-only mode. As a reminder, today's call is being recorded.

I would now like to turn the call over to Steve Haymore from First Solar Investor Relations. Mr. Haymore, you may begin..

Thank you, Justin. Good afternoon, everyone, and thank you for joining us. Today, the company issued a press release announcing its fourth quarter and full year 2016 financial results. A copy of the press release and associated presentation are available on the Investors section of First Solar's website at firstsolar.com.

With me today are Mark Widmar, Chief Executive Officer; and Alex Bradley, Chief Financial Officer. Mark will provide a business and technology update, then Alex will discuss our fourth quarter and full-year financial results and provide updated guidance for 2017. We will then open up the call for questions.

Most of the financial numbers reported and discussed on today's call are based on U.S. Generally Accepted Accounting Principles.

In the few cases where we report non-GAAP measures such as free cash flow, adjusted operating expenses, adjusted operating income or non-GAAP EPS, we have reconciled the non-GAAP measures to the corresponding GAAP measures at the back of our presentation.

Please note, this call will include forward-looking statements that involve risks and uncertainties that could cause actual results to differ materially from management's current expectations. We encourage you to review the Safe Harbor statements contained in today's press release and presentation for a more complete description.

It is now my pleasure to introduce Mark Widmar, Chief Executive Officer.

Mark?.

Thanks, Steve. Good afternoon, and thank you for joining us today. While market conditions and pricing remain challenging in the fourth quarter of last year, we finished 2016 with strong results. An important part of our DNA at First Solar is to set challenging goals and hold our-self accountable to them.

At our guidance call in December of 2015 and as a part of our Analyst Day in April of 2016, we provided operational and financial metrics for investors to measure our progress against. As Alex will review later, we have been able to meet and even exceed these targets. First Solar's ability to deliver on commitments also extends to our multiyear plans.

For instance, at our 2013 Analyst Day, we outlined a goal to achieve an exit efficiency of 16.9% at a module cost per watt of $0.45 by the end of 2016. Nearly four years later, our best line exited 2016 at over 16.9% conversion efficiency and our module cost per watt for the year beat our target by a wide margin.

These are remarkable achievements that demonstrate the expertise and execution capabilities of the First Solar team, and give us confident as we again set challenging goals related to the Series 6 program.

Since making the decision to accelerate our transition to Series 6, the First Solar organization has been completely focused on executing to our planned roadmap. These efforts fall under the areas of both product and market readiness.

While we'll be covering our product readiness efforts on today's call, we have been actively engaged with customers and ecosystem partners to educate them and solicit feedback from them on this powerful new product. We are very pleased with the initial customer reaction in market readiness assessment.

We will provide more details related to this ongoing work on our Q1 call. Turning now to slide 4, I'll review the efforts and progress we're making to ensure our Series 6 module meets the performance, cost and production targets we have established. Firstly, it is important to put in proper context the risk of this transition.

While there're undoubtedly challenges involved, there are several key factors that reduce the overall risk profile and give us confidence in our ability to successfully deploy Series 6. Most significantly, our Series 6 module will utilize essentially the same underlying solar cell technology as our Series 4 product.

Given this, we believe the core technology risks involved in the transition is low. The analogy we have used in the past is that this is similar to what and how the flat panel display industry scale from one generation to the next. We are not reinventing the core technology behind Series 6, but rather increasing the module form factor.

In addition, we're deploying this new technology with experienced R&D personnel, deploying it in our existing factory locations with manufacturing teams that have collectively produced over 17 gigawatts of CadTel modules.

The cumulative years of experience, working with this technology within our organization, provides us with the skill set required to successfully launch and ramp the Series 6 production platform.

In a similar matter, but external to the company, we are leveraging the expertise of the long-trusted equipment supplier for our core product technology tools. In some cases, we have been working with these suppliers for nearly a decade and their capabilities are well established and proven.

While there clearly are risks associated with scaling to a larger form factor, we see these risks as manufacture-related in nature and not technology centered.

While manufacturing risks such as throughput and yield could possibly impact the Series 6 launch, our experience indicates that overcoming these types of challenges can be addressed with a proper focus and resources.

From a performance standpoint, we are focused both on reaching our target of greater than 18% efficiency as well as continuing our existing standard of excellence and reliability that our customers have come to expect.

While our target efficiency represents a greater than 100-basis-point improvement from our current fleet performance, the path to achieve this performance is largely based on proven and well-understood methods that we will now be able to deploy on our new Series 6 equipment addressing certain limitations in our existing toolset.

One portion of the efficiency improvement comes from the scaling of our form factor, which increases the active area of glass relative to the total area. Another efficiency improvement comes from changes to electrical design of the module, a process that we have experience with and which is well proven.

The remaining efficiency improvement will be designed into certain process tools, where we'll again leverage the expertise of our longstanding equipment suppliers.

On the quality and reliability side, we are applying our same change management systems, which has enabled us to produce multiple gigawatts of world-class modules, while driving up efficiencies over the past number of years.

Intensive validation of the elements of our Series 6 design has already begun, including among other methods, utilizing accelerated life testing and field exposure testing. Ensuring we reach our target module cost per watt in Series 6 is another critical aspect of our product readiness efforts.

One area we have made tremendous progress is our CapEx per watt target. We have now solidified our ability to meet and even potentially do better than our $0.30 per watt CapEx target for brownfield capacity. This has been achieved through negotiations with our suppliers and by expanding our supply base for non-critical process tools.

Regarding the factory labor component of our cost per watt, we have worked through extensive evaluations of our labor requirement and determined that we are also on track.

As relates to the material cost of the module, we will continue to negotiate with both existing and new suppliers to further drive down material costs and enable us to achieve our targeted cost profile. Taken all together, we view the challenging cost per watt target we have set for Series 6 as appropriate.

Turning to slide 5, I'll review some of the key milestones that we are providing in order to evaluate our progress of the Series 6 deployment. Note, the updates currently provided are limited to our initial pilot line in the Ohio factory and the first full scale production facility in Malaysia.

In subsequent quarters, we will add key milestone for the additional production facilities included in our manufacturing roadmap. Near the end of last year, we stopped production of four lines at our Ohio factory and began preparation for the installation of our Series 6 pilot line.

At this point in Q1, we have placed orders for the large majority of the tools including all the major equipment required for the initial pilot. Over the course of the coming months, we will work closely with our suppliers to factory test and validate the tools before the equipment begins to arrive at our Ohio location in the third quarter.

By the end of the third quarter, the core Series 6 tools, which include equipment such as our coders are expected to be operational. We expect to complete the frontend of our pilot line in the fourth quarter, with production target started in the second quarter of 2018.

Our Ohio plant is expected to have a nameplate capacity of 550 megawatts when fully ramped by Q4 of 2018. With regards to our Malaysia factory, we will be stopping Series 4 production on eight lines within the next two months. Tool ordering for Malaysia production has already commenced, and will continue into Q2 of this year.

Production startup in Malaysia is targeted for Q3 2018 through Q4 of 2018. Once fully ramped by Q1 of 2019, this portion of Malaysia factory is expected to have approximately 1.1 gigawatts of Series 6 capacity. While this is the first look at our key milestones, we'll continue to update and add additional relevant information as needed going forward.

Turning to slide 6, I'll focus on some of the bookings activity since our third quarter earnings call last November. Our most recent bookings are highlighted by encouraging progress in the Asia-Pacific region. In Australia, we reached a new milestone with the award of the PPA for our first self-developed project in the country.

The approximately 50-megawatt AC Manildra Solar Farm, which was awarded under a grant under the ARENA program's large solar scale program. We'll begin shipping module to the project in 2017 with the expected completion in 2018. In addition, we have recently signed two separate module supply agreements, which combine total over 200 megawatts DC.

The 140-megawatt DC supply agreement with Sun Metals will provide energy to the company's zinc refinery and will be the largest solar power plant in Australia once completed. In a separate transaction, we will be supplying 63 megawatts DC for the first phase of the Kidston Solar Project, which will be co-located with the Pumped Storage Project.

Note, Series 6 will be well-positioned for the Kidston's potential second phase. Shipments to both projects are scheduled in 2017. We're enthusiastic about our recent success in Australia and the growth potential of the large-scale utility solar market in the region, where our technology hold a strong energy advantage.

In Japan, we closed our first non-recourse syndicated project financing led by Mizuho Bank, one of the largest financial institutions in Japan, for our 59-megawatt project in Ishikawa. The financing arrangement demonstrates First Solar's technology, project development and project financing capabilities for utility-scale solar projects in Japan.

Additionally, we have booked two new development projects with a combined size of nearly 50 megawatts DC. Module shipments to the two project will take place over the course of 2017 and 2018. These bookings bring our contracted Japan development pipeline to over 180 megawatts DC, including 10 smaller projects which have commenced operations.

We're encouraged by the future opportunities we have in the market, based on our mid- to late-stage bookings opportunities that now total over 350 megawatts DC. In the United States, we signed an agreement with a major ITP to supply modules to a 200-megawatt DC power plant in the Desert Southwest.

Module shipments to the project are also scheduled for this year. The remaining bookings for the quarter were primarily module-only sales in India, Turkey and other parts of Europe. As it relates to Turkey, we recently announced a collaborative sales agreement with Zorlu Holdings.

Under the five-year agreement, Zorlu has the right to distribute CadTel technology in Turkey and 25 other countries, primarily in Southeast Europe, the CIS and Central Asia. The agreement will support Zorlu's distribution efforts as well as bolster our indirect sales model.

As part of the agreement, First Solar's business development team in Turkey will transition to Zorlu. In recent years, our strategy has been to dedicate sales and operational resources to select markets that are sufficient scale, and with sufficient long-term sustainable growth to support direct OpEx.

In other parts of the world, where there is demand for solar, but the market opportunity requires a greater local presence, our indirect model with partners such as Zorlu and Caterpillar offer an effective and OpEx-like solution. We'll provide more updates in the future on our collaboration progresses.

Returning to our recent bookings, since the last quarter, we have booked over 650 megawatts DC. For the 2016 calendar year, we booked 1.8 gigawatts with additional bookings of over 400 megawatts so far this year. While our book-to-bill ratio for 2016 fell below our target of 1 to 1, this is large part due to the challenging ASP environment.

Module ASPs began to decline significantly in July of 2016. As we observed the declines, we engaged the market with a price discovery approach to determine the market clearing price. During this period of time, we were out of price position, which is reflected in the relatively low booking volumes in the second half of 2016.

We took actions to correct this, based on the observed market clearing prices, and this is reflected in the bookings momentum we have seen since our last earnings call. Netted against these recent bookings is the removal of our pipeline of a 310-megawatt AC Tribal Solar development project, which was awarded in an RFP process in 2014.

In light of significant uncertainties and risks related to land use rights, we have discontinued the development of this project. As it relates to our remaining contracted development pipeline, we view this as an isolated event based on risks specific to this project.

While this decreases our volume of contracted shipments in the 2020 and 2021 timeframe, we see opportunities to offset this impact as in the case of our 350 megawatts of potential bookings in Japan previously discussed, and leveraging the strength of our Series 6 product.

Net of the reduction of Tribal Solar, our remaining contracted shipments now stand at 3.3 gigawatts DC. Turning briefly to slide 7, our remaining 3.3-gigawatt of contracted volume equates to $5 billion of future expected value as of today's call.

The decreases in expected value over the course of the year is due to the higher mix of systems business recognized in 2016 as compared to the higher mix of third-party module bookings during the year. The remaining $5 billion total on this slide is also net of future expected value of Tribal Solar.

Note that, going forward, we intend to discontinue the use of this metric. Near term, majority of our bookings will be focused on selling through the remaining Series 4 volume, which is anticipated to primarily be module-only sales.

Longer term, as we grow our production platform by leveraging the strength of our Series 6 product, we will see a shift to more module-only sales. Given the book-to-bill velocity of module-only sale is much shorter than a system development project sale, the inherent value of a multi-year forward-looking contracted pipeline is diminished.

This is consistent with our prior statement that the system business will reflect approximately 1 gigawatt of annual volume with the balance, approximately 2.5 gigawatts, based on 2019 anticipated production being module-only.

An important new metric that we're providing on slide 8 is the sell-through status of our expected remaining Series 4 production. The first thing to keep in mind is that the 3.6 gigawatts to 3.8 gigawatts of the Series 4 supply is representative of both current inventory and expected future production into 2018.

Depending on certain factors, including market demand, module pricing and the progress of our Series 6 ramp, we may decide to adjust the ramp downtime of certain Series 4 lines, which could impact Series 4 supply by up to 200 megawatts.

Note, transitioning resources earlier in 2018 from Series 4 to Series 6 will allow for a more efficient launch in our Ohio facility. Again, the total forecasted Series 4 supply approximately 600 megawatts which will be allocated to our own project pipeline with additional 1.4 gigawatts contracted for delivery to third-party customers.

These amounts are inclusive of bookings we've discussed on today's call. With over half of the project volume already contracted, we're making good progress in selling out of our Series 4 supply.

In addition to the 2 gigawatts already contracted, we have a number of mid-to-late stage opportunities that could be contracted against the remaining supply as shown on slide 9. Note, approximately 50% of this volume is in late-stage negotiations.

Previously, we've provided a view of our entire portfolio of potential bookings including early-stage projects. In this year, we are focusing on only those mid-to-late stage opportunities, which have the greatest likelihood of bookings.

We continue to track a large number of early-stage opportunities and while the total has increased since our last earnings call, we feel this targeted view is more relevant as we transition from Series 4 to Series 6.

Of the opportunities highlighted, approximately 1.7 gigawatts of our Series 4 delivery with the remaining 500 gigawatts of opportunity is associated with Series 6. Again, there exists a much larger pipeline of early-stage opportunities, but this view is only for those in mid-to-late stages.

Of the 2.2 gigawatts of potential bookings, we continue to have the most opportunity in the U.S. across both development and module sales. Most notably, there are over 500-megawatt DC of module sales opportunities in the U.S. that are in late-stage negotiations, and once booked, would reduce remaining Series 4 supply.

APAC includes the Japan pipeline mentioned as well as additional opportunities in Australia and parts of Southeast Asia. In Europe, the majority of the opportunities shown are in France. India continues to be an important market.

And while our higher profitability opportunities are strong, we have a much larger number of early-stage projects in this market. Lastly, while we have historically held our Analyst Day event in the spring of each year; this year, we will be changing the time of the meeting to the fourth quarter.

Later this year, we will be further into our Series 6 transition and the revised timing allows us to have a more meaningful information to share. We'll provide more details on the timing and location of the event at a future date.

Alex will now provide more details on the fourth quarter and 2016 financial results, and discuss updated guidance for 2017..

Thanks, Mark. Before reviewing the financial results for the quarter, I'll turn first to recap our accomplishments in 2016. In terms of efficiency, we met the targets we outlined in April at our Analyst Day. Our 2016 full fleet efficiency of 16.4% is an 80-basis point improvement versus 2015 and a remarkable 320-basis point improvement since 2013.

Our full fleet exited the year at 16.7%, also meeting the target we set in April. In addition to achieving our efficiency target for 2016, we also met and even exceeded the module cost per watt target that we set. While we did not disclose that number, we beat the 2016 cost goal and our cost per watt decreased by 16% compared to full-year 2015.

From a financial perspective, we also delivered strong results for the year. The initial 2016 earnings per share midpoint we provided in late 2015 was $4.25 per share. In November, we raised our earnings midpoint to $4.70 per share.

Our non-GAAP or operational earnings of $5.17 per share for the year exceeded both our original and revised guidance for EPS. While our revenue and net cash came in below the initial guidance provided, that's a result of revised timing of certain project sales.

So, in summary, we delivered on our commitments this past year and continue to apply the same discipline and focus to the objectives we've outlined for the coming years. Beginning on slide 12, I'll highlight the operational achievements for the past quarter.

Fourth quarter module production was 760 megawatts DC, a decrease of 2% from the prior quarter, due to the production stop on certain lines in Ohio related to our Series 6 transition. The ramp down of these lines also impacted our capacity utilization, which decreased to 92% in Q4 versus 97% in the third quarter.

Capacity utilization was 100% in the same quarter of 2015 as all lines are operating with minimal efficiency upgrade activities. Our conversion efficiency for our full fleet averaged 16.6% in the fourth quarter, which was an increase of 10 basis points quarter-over-quarter and a 50-basis point increase year-over-year.

Module conversion efficiency on our best line improved to a Q4 average of 16.8%, a 20-basis-point improvement versus Q3. Our lead line exited Q4 at 16.9%, which was unchanged from the prior quarter. Turning to slide 13, I'll next touch on some of the income statement highlights for the fourth quarter and full year.

Note that in the fourth quarter of 2016, we adopted a new accounting standard for the treatment of share-based compensation, which resulted in changes to tax expense, operating cash flows and financing cash flows.

The following net income and earnings comparisons to prior quarters are also reflective of this change and more information related to the adoption of the new standard will be available on our 10-K.

In addition to reviewing our income statement results, I'll be discussing certain non-GAAP measures such as adjusted operating expenses, adjusted operating income and non-GAAP earnings per share. Please refer to the appendix of the earnings presentation for the accompanying GAAP to non-GAAP reconciliations.

Net sales were $480 million, a decrease of $208 million compared to the prior quarter. The decrease in net sales resulted from the completion of the Taylor, East Pecos, Astoria and Butler projects in the quarter.

The lower revenue from these projects in Q4 was partially offset by the sale of our Shams Ma'an project in Jordan and higher third-party module sales. For 2016, net sales were $3 billion as compared to $3.6 billion in the prior year.

Also keep in mind that the Q4 sale of our remaining interest in the Stateline project to 8point3 for $280 million in cash and a $50 million promissory notes was not accounted for as revenue, rather profit on this sale was recognized in equity in earnings.

As a percentage of total quarterly net sales of solar power systems revenue, which includes both our EPC revenue and solar modules used in systems projects, decreased from 69% in the prior quarter to 48% in Q4, resulting from the higher third-party module sales and completion of systems projects mentioned.

For the full year, solar power systems revenue was 77% of total revenue. Gross margin for the fourth quarter was 13% compared to 27% in the prior quarter.

The decrease in gross margin percentage was a result of the mix of projects recognized between the periods, and a $25 million non-cash impairment of our Barilla Solar Project in Texas, which is classified in PV solar power systems on our balance sheet. For 2016, gross margin percentage was 24% compared to 26% for the prior year.

The project impairment had an approximately 500-basis point and 80-basis point impact on Q4 and full-year 2016 gross margin respectively.

The 30-megawatt Barilla project was originally developed to sell power in Texas on an uncontracted basis in order to help penetrate the Texas market as well as to provide a test site for the implementation of new technologies.

We've seen benefits from this effort in the form of our successful development sale of the East Pecos project along with the successful launch of our 1,500-volt inverter infrastructure.

However, declines in retail power prices since the completion of the project in 2014 have resulted in ongoing operational losses that necessitated the write-down in value. Gross margin of our components segment was 16% in Q4 compared to 32% in the prior quarter. The decrease is primarily due to the Barilla impairment and lower third-party module ASPs.

Operating expenses, excluding restructuring and asset impairment charges, were $100 million in Q4. This compares to Q3 adjusted OpEx of $93 million. Operating expenses for the fourth quarter included approximately $8 million for the impairment of development costs associated with the Tribal Solar project.

For the full year, adjusted operating expenses were $388 million. Restructuring and asset impairment charges to accelerate our Series 6 transition was $729 million and $819 million for the fourth quarter and full-year 2016 respectively.

These charges were primarily non-cash and the Q4 charges were higher than the anticipated restructuring charges of $500 million to $700 million provided during our guidance call in November as well as to our initial estimate impairment charges in our Series 4 manufacturing lines were higher than anticipated.

Note also that charges of $40 million to $60 million related to the cancellation of our Series 5 operations, which were included in this range, are now expected to be incurred in 2017. Excluding restructuring related items, we had an operating loss for the quarter of $34 million compared to adjusted operating income of $98 million in Q3.

The decrease is primarily due to the Barilla asset impairment, projects impairments, lower revenue and the mix of projects recognized. Keep in mind that the $125 million profit on the sale of the Stateline project is not included in operating income.

On a GAAP basis, our operating loss for the quarter was $765 million and the 2016 operating loss was $503 million. Other expense was $8 million in the fourth quarter, primarily due to the impairments of a cost method investment. We had tax expense of $90 million in the fourth quarter compared to tax benefit of $66 million in Q3.

The tax expense included $196 million associated with the distribution of approximately $750 million of cash to the U.S. from a foreign subsidiary. Of the $196 million tax expense, only $7 million is expected to result in a cash payment. For the full year, our tax expense was $58 million compared to a benefit of $6 million in the prior year.

Full-year tax expense was impacted by the tax from the distribution of cash, partially offset by a $35 million tax benefit in Q3 from a favorable ruling from its foreign tax authority.

Equity in earnings was $146 million in Q4, primarily composed to the profit on the sale of our remaining interest in the Stateline project and earnings from our investment in 8point3. This compares to $11 million of equity in earnings in the third quarter, which was comprised primarily of equity in earnings from our investment in 8point3.

For the full year, we recognized $172 million of equity in earnings compared to $20 million in the prior year. Altogether, in the fourth quarter, we recognized the loss of $6.92 per share on a GAAP basis and earnings to fully diluted share of $1.24 on a non-GAAP basis. This compares to GAAP earnings of $1.63 in the prior quarter.

For 2016, our loss per share on a GAAP basis was $3.33 and earnings per share was $5.17 on a non-GAAP basis. This exceeds the $4.80 high-end of EPS guidance we provided on our last update and significantly higher than the $4.50 high-end of EPS guidance in our original 2016 guidance.

Relative to our non-GAAP EPS guidance, our results exceeded our guidance as a result of higher sales value for the remaining interest in the Stateline project and lower taxes.

At the time of our guidance update in November, the final structure of the Stateline sale to 8point3 had not received final board approval, and a portion of the sales proceeds were not fully reflected in our guidance. I'll next discuss select balance sheet items and summary cash flow information on slide 14.

Our cash and marketable securities balance was slightly below $2 billion at the end of 2016, and decreased to $135 million from the prior quarter. Our net cash position improved by $464 million to nearly $1.8 billion, as we had strong cash receipts in Q4. We also paid off our borrowing on our revolver during the quarter.

In Q4, our net working capital, which includes the change in non-current project assets and excludes cash and marketable securities, increased by slightly more than $450 million. The change was primarily related to repayment on the revolver and higher project assets, partially offset by an increase in other liabilities.

Total debt was $188 million in the fourth quarter, a decrease of $599 million from the prior quarter. The decrease primarily resulted from $550 million repayment of our borrowing under our revolving credit facility and the partial repayment of a project-related VAT loan.

Cash flows from operations were $268 million in Q4 compared to cash flows used in operations of $84 million in the prior quarter. Free cash flow was $215 million compared to negative free cash flow of $130 million last quarter. Capital expenditures were $54 million as compared to $46 million in the prior quarter.

For 2016, operating cash flows were $207 million. The cash generation of the business was very strong this past year, particularly when taking into account a couple of different factors.

Firstly, the $280 million received this year from the sale of our remaining interest in Stateline was treated as an investing cash flow and is not reflected in the operating cash flow for the year.

In addition, we largely constructed multiple projects this year, such as Moapa and the first phase of California Flats, but received only a partial payment on Moapa. We exited the year with a stronger balance sheet and cash position as we've ever had and are well positioned for our Series 6 transition.

I'll next discuss updates to our full-year 2017 guidance on slide 15. Before delving into the details, there are certain key assumptions underlying our guidance to keep in mind. Firstly, we're likely to early adopt the new revenue standard in Q1 2017, which provides guidance on recognizing revenue from contracts with customers.

As it relates to our 2017 net sales guidance, the new revenue standard is not expected to have a significant impact. However, as in the past, the final structuring of project sales can impact our outlook. Secondly, we made significant progress in the sale of our Moapa project with the recent closing of the tax equity sale.

Additionally, an agreement to sell the cash equity from the project has been signed with final closing expected in March upon receipts of third-party approvals. Based on the final structuring of the Moapa transaction, the sales will be combined and accounted for as a single arrangement.

Accordingly, we now expect to recognize the full revenue on the project, which adds approximate $0.3 billion to our guidance range. As we've indicated in the past, the structuring of project sales can result in different accounting outcomes.

Due to the uncertainty of the final deal structure, our guidance provided last November did not include the full amount of revenue, but did include the full gross margin. Therefore, this guidance change results in increased revenue without any increased gross margin.

Thirdly, as it relates to our California Flats project, we anticipate selling the project this year, but the timing is uncertain. The transaction structuring has been driven with the aim of providing a residual interest to be offered to 8point3, which is added to the deal complexity.

Maintaining a residual interest to 8point3 continues to be our aim, but given market uncertainty around potential implications of tax reform, the deal is taking longer than usual to structure. Turning now to our updated guidance ranges, note first that on a non-GAAP basis, we're not making any changes to our ending earnings per share expectations.

Our new net sales range, adjusted for the Moapa transaction structuring, is $2.8 billion to $2.9 billion. As a result of the increased revenue and no incremental margin dollars, the gross margin percentage range has been updated to 11% to 13%. Non-GAAP EPS remained unchanged at breakeven to $0.50.

In terms of the quarterly distribution on non-GAAP earnings in 2017, we expect a loss of $0.10 to $0.15 in the first quarter of 2017 and to be in an overall loss position in the first half of 2017 with earnings weighted towards the second half of the year.

Key drivers of the first quarter loss position include lower module shipments in the first half versus second half of the year, as well as a higher mix of systems revenue and gross margin expected in the second half of the year.

Also note that despite being in a loss position, the Q1 loss per share includes an assumed tax expense due to the jurisdictional mix of income. Turning next to our GAAP operating guidance, the range has been revised to $335 million to $380 million, which includes $55 million to $80 million of cash restructuring-related charges.

The range includes $40 million to $60 million of charges related to the cancelation of our Series 5 operations, the timing of which was originally anticipated in 2016. In addition, we've revised the range for expected severance and other charges to $15 million to $20 million, reflecting revised timing of charges originally expected in 2018.

As a result of the updated operating expense, we've revised our expected operating income range to a loss of $40 million up to $25 million of income. The revised net loss per share range of $0.80 to $0.05. Turning to balance sheet, we're maintaining our ending expected net cash balance range of $1.4 billion to $1.6 billion.

Although our ending 2016 net cash came in above our guidance, this was primarily a result of higher-than-guided operating cash flow in 2016 as certain payments and receipts came in ahead of expected schedule. We expect this to reverse out in 2017 operating cash flow. And for this reason, our expectation for net cash at the end of 2017 is unchanged.

The impact of the previously mentioned timing of receipts and payments results in lower operating cash flow expectations for this year. As a result, we are revising the range to $250 million to $350 million from the prior range of $550 million to $650 million. And capital expenditures and shipments remain unchanged from the prior ranges provided.

I'll summarize our progress in the fourth quarter and full-year 2016 on slide 16. We had strong financial results for the year with net sales of $3 billion and non-GAAP EPS of $5.17, exceeding the high end of our guidance. Our ending cash was nearly $2 billion with $1.8 billion of net cash. For 2017, we're maintaining our non-GAAP EPS of zero to $0.50.

Our module efficiency for the year and for Q4 was impressive with a full fleet average of 16.4% and 16.6%, respectively. Our best line efficiency exited the year running at 16.9%. And lastly, our 2016 bookings total was 1.8 gigawatts and we booked over 400 megawatts thus far in 2017.

With 2.2 gigawatts of mid-to-late stage opportunities, we'll continue to have healthy bookings prospects. And with that, we conclude our prepared remarks and open the call for questions.

Operator?.

Well, thank you. Our first question comes from Paul Coster with JPMorgan..

Yeah. I just want to check on Moapa. The $300 million in revenues that have been added in, is not actually a change to the economics, it's just merely an accounting adjustment as a function of how you've negotiated the tax equity arrangement. And then I've got a quick follow-up..

Yeah, that's right. So, there's no change to the economics. It's just an accounting change due to transactions, tax equity and cash equity now recognize both together. So, it just creates higher revenue, but doesn't change the overall economics..

All right. And then the 2017 guidance, back-end loaded year. Can you just give us a little bit of color behind that? What gives you the confidence in the second half ramp? And whether the gross margins will be sort of constant or whether they inflect as well? Thank you..

Yeah. I'll take, I guess, on the guidance side.

So, Paul, I think one of the things that we will highlight is that the bookings momentum that we're starting to see now, and again as we highlighted in the prepared remarks that, there was a period of time that we were practically out of price position, doing a little bit of price discovery in the marketplace to really understand where the market clearing price was going to be.

I mean, it was a very significant disruptive decline in module ASPs and starting in July. And what that did is, is resulted in lower bookings momentum really through towards the latter half of 2016. And we've seen significant momentum moving forward, we just recently booked since the beginning of this year over 400 megawatts.

And so, what that means is, this going to position a lot more volume, the module shipment volumes that we had anticipated is going to be positioned towards the second half of the year. So, part of it is that.

The other piece is, is that, the current profile of the timing of the revenue recognition on our systems business will be more heavily weighted towards the second half of the year. So, I have confidence in terms of our profile from the standpoint of the book business that we have right now.

We have a very strong booked module volume at this point in time, given that what we've now been able to book over the last, call it, four, five months, plus what we have contracted now for the systems business and where we are and in that side of the house, we feel very confident with.

Now, all that what I would say is, the wildcard that could influence that will still end up being the sell down of our systems business. We're largely on our way with our Playa negotiations right now on our switch project. CA Flats, we're still moving forward with that.

And so, the timing of that could impact the second half volume ramp that we're anticipating, but again that is just a movement of an economics that could shift from third and fourth quarter or potentially fall into the beginning of 2018, but as always said, the system business can be little lumpy.

It doesn't impact overall economics, it's just timing associated with that. And like I said, is there some risk to that, but other than that, we feel highly confident with the balance of the year forecast..

And our next question comes from Brian Lee with Goldman Sachs..

Hey, guys. Thanks for taking the questions. Just had a couple. I guess, first, Mark, on Tribal Solar, SCE was already a PPA owner. So, wondering if you guys had done any work on trying to re-strike that PPA. And then, if you look at the list, there's 650-or-so megawatts of other SCE projects for 2018 to 2020.

So, any update you can provide on the status of those, any of those maybe potentially having similar risk to what you saw with Tribal Solar. And then, I had a follow-up..

Yeah. So, we were in active and ongoing discussions with SCE as it relates to Tribal Solar. The issue is not with SCE. The issue is ultimately with the tribe and their desire for completion of the project on tribal land. Initially, they granted the option and their consent for the construction of the project on their reservation.

Because of cultural issues and evolution of changes of certain leaders within the tribe and momentum from the balance of the constituents, there was a change in the support for the project. We tried to resolve those issues. We could not successfully do that.

We had also had incurred a relatively small increase to the original cap on the network upgrades. We effectively used that. Given we were unsuccessful in our ability to influence the tribe to support the project, and without their support, we would not be able to complete the project.

We effectively use that provision under the cap of the interconnection agreement effectively or the upgrade to the network to result in a termination of the PPA. It's a unique situation. We don't have any other similar situations. We've done other projects with, for example our Moapa project was on tribal land.

We had no issues working in with the tribe, the Moapa tribe. The current situation that we had there was more of a challenging environment, and it resulted in unfortunately the termination of that PPA.

The balance of the projects that we have with SCE, none of them are on tribal land, and we have full commitment and support with SCE for those projects..

Okay. Great. That's helpful. Just a second question, and I guess a little bit similarly on Barilla. Maybe a question on timing. Why write it down now? It's been, my understanding, operating for some time in the current low power pricing environment. So, would just be curious on why it wasn't written down sooner. Thank you..

Yeah. So, Brian, I mean, we've been following the project and we've now, we feel, got enough operational data to make a better determination of long-term prospects for the project. So, Barilla was originally developed to penetrate the Texas market and to be a test like for new technologies.

We constructed it with lower bin module, so we had a higher installed cost there than we might otherwise have had, and it's given us some benefits around our ability to win the East Pecos deal, and launch of our 1,500-volt architecture.

But we looked at it now and based on where we see declining spot pricing at the moment and given the recent operational edge, we felt now is the right time to write that down..

Yeah. I think in the accounting world, Brian, you may know, I mean, there's obviously the triggering events. And one of them is, do you have a potential impairment and is that impairment other than temporary, right.

And even though there was indications through there are the operations of the asset, at that point in time, it was unclear whether or not the potential risk of an impairment was going to be permanent in nature, and what has ended up happening given where the current power prices are and how we see that evolving in the near-term, it trigged an event that says, yes, now we believe that the impairment is other than temporary.

It also is, as Alex indicated is that, we did use low bin modules. This was really was viewed more as an R&D in endeavor, and to use that as an opportunity to continue to test new products and whether it's modules, whether it's inverters, whether it's other components within the architecture, we'll continue to use that site for that.

So, there inherently will be value there. So, I look at this as, it's mainly just it's an accounting item. We'll continue to sell the power that's being generated off that asset, but from a book value standpoint, we had to write it down..

And our next question comes from Tyler Frank with Robert Baird..

Hi, guys. Thanks for taking that question. Can you talk a little bit more about the overall marketplace? You made some comments that you were out of position in terms of pricing in the back half of last year.

So, what sort of adjustments did you have to do in order to start getting bookings in the first half? And should we expect modular sales to have extremely low margins based on the current market pricing? And how should we think about each individual market in terms of ASPs currently?.

First off, as it relates to each market on an ASP basis, I would argue that the relative baseline of crystalline silicon prices relatively consistent globally.

But again, where we can capture better value is selling into markets where we have an inherent energy advantage and capture the value of that energy that's being generated and we can price at a premium.

So, if you look at the bookings that we've recognized this quarter, I would say, there's a number of them, module-only sales that we were able to capture reasonably meaningful premium to where crystalline silicon is pricing at this point in time, because we've sold the products into geographies where there is inherent energy advantage and we capture that.

When you saw the module pricing decline as rapidly as we did and starting in July, I think it's very prudent to go and continue to test it and to see where the prices are starting to settle out at. We have started to see prices settle a little bit.

They seem to be in a relatively consistent range when we think about not only within the U.S., but globally. And so, we've just adjusted to that market pricing and we're continuing to go on and sell the value of the energy and we've been successful in doing that and starting to show up in our bookings.

Now, granted, yes, with the margin that we're realizing on the Series 4 product be at a lower margin than we would look to as a long-term entitlement, clearly it should, because as we've highlighted, the reason we're transitioning into Series 6 is because of the smaller form factor of Series 4, which in different regions of the world can result in a BoS penalty that could be in the range of $0.06 to $0.08.

So, that's a meaningful delta, plus Series 6 comes at a much lower cost profile. And as we highlighted in our last call that Series 6 will have a profile that's in the range of 40% lower than Series 4. So, when you capture the energy yield advantage, slightly higher efficiency product in Series 6 versus Series 4 at a much lower cost profile.

Well, yes, it's a very challenging market, having a Series 6 product in this market environment, we couldn't be better positioned..

And our next question will come from Krish Sankar with Bank of America..

Yeah, hi. Thanks for taking my question. Mark, I just want to follow-up on one of your comments you mentioned on the module pricing. You said it's stabilized. Do you feel like the module pricing for the industry is bottoming out, or do you think there's another leg down? And then, a follow-up on the cost profile, Series 4 versus Series 6.

If I remember right, Series 4 had about $0.07 per watt of depreciation cost. What do you think it's going to be for Series 6? Thank you..

So, look, I think the relative stability of the market is always going to be determined by supply and demand.

And there's obviously triggering events similar to what we saw in the 2016, whereas it relates to the second half of 2016 where there was a pretty significant disruption of demand, primarily because of tremendous build out of solar in the first half of 2016 in China. That risk profile will exist on a perpetual basis.

Whenever we get into an imbalance of supply, demand to the extent a particular geography that is a meaningful component of the overall global demand perspective, starts to see a shift or a decline, then they always are going to be subject to these volatile times, and ASPs can change very quickly.

What we're seeing right now is a relatively stable environment, but it could change very quickly similar to what it did in 2016. So, we'll have to keep a watch on that.

And what we have said before, our assumption in long term is that this perpetual oversupply will exist, that our competitors will continue to sell at or below cash costs in order to run their factories as efficiently as possible.

And we just have to create a product platform, which is why we're transitioning to Series 6 that will enable us sustain that very challenging market environment; sell the value of the energy, capture the cost entitlement of the product and compete in, even though what some would argue, is an unsustainable market environment.

It may be for some of our crystalline silicon competitors, but we need to create a business model that can sustain that type of environment. As it relates to the depreciation for Series 6 versus Series 4, I'll let Alex take that question. So, anyway, I guess, Alex didn't hear the entire question.

So, on Series 6, the depreciation, you could look at it from the standpoint of the CapEx is approximately half of Series 4. So, from a greenfield standpoint, and given that we are actually doing a brownfield expansion for Series 6, we're going to see a lower depreciation expense.

We haven't given the exact number, but you can envision that it will be at a lower CapEx – or depreciation expense, excuse me, than the Series 4 product, just largely because the CapEx per watt is lower..

Yeah, Krish. Apologies for not hearing that, but if you look to the numbers we gave in the guidance call we gave in November, we guided to a CapEx per watt there. And you can look at that relative to the numbers we've given historically around Series 4, and look at those two in find a relative number on depreciation..

And our next question comes from Vishal Shah of Deutsche Bank..

Hi. Thanks for taking my question. Mark, just on the systems business.

Some of the challenges that you're hearing and seeing in the marketplace today with respect to the corporate tax reform, how is that impacting or could impact margins in the second half of the year? Is that incorporated in your guidance? And as we think about Series 6, I know you had said low- to mid-$0.20 per watt cost targets.

It looks like your CapEx is better than what you were planning before. So, are we looking at the low $0.20s as a range for your Series 6 cost target? Thank you..

Yeah. I'll take the Series 6 cost and I'll let Alex take the tax administration discussion. So, yes, I mean, we're very happy with where we're trending right now from a CapEx perspective. As you would envision, there are many, many different variables that ultimately are going to impact the final cost profile for Series 6.

Throughput yield will impact that as well. A lot of things that we have to continue to move forward. We're very encouraged with what we're seeing so far, and obviously having a lower CapEx per watt is obviously a very good indicator of the opportunity set.

I would say the challenge that we have right now, the one that we need to stay as aggressively focused on as possible for Series 6 is really going to be our bill of material cost. So, we need to continue to drive down our bill of material cost. We have our roadmap to make that happen. There is a lot of work, though, to ensure that we can get there.

And we're going to have to work aggressively in leveraging and negotiating and partnering with our suppliers to ensure that we can do that. So, encouraged by the CapEx early indications, but obviously a long way to go to actually achieve and potentially even do better than our expectations around the Series 6 cost profile..

And, Krish (sic) [Vishal], as it relates to tax, so I mean, firstly, our guidance today is based on current tax policy, and assumes no changes to that. If you look at it on a project level, the two key drivers to value there are going to be ITC and depreciation. On the ITC side, I think it's unlikely we'll see any change.

If you look at the last time the extension came through, it was supported strongly on a bipartisan basis for that extension. Renewables has been responsible for a pretty significant job creation and it already has a finite term. So, if you look at history around tax credit elimination, normally you see a transition period.

So, we would expect that there'll be no change to the current ITC schedule. On the depreciation on makers side, who knows what that change will be. It's hard to estimate, but we expect any change there would be offset by a corresponding change in tax rates.

What I will say from a structuring perspective is that the uncertainty does create some challenges. There are players in the market who are still open to doing business.

So, we don't see an issue with getting tax equity on deals, but there may be changes to the amount of tax equity going into deal to the structures and perhaps to the risk profile the tax equity is looking at in the short term. But we don't see any issue with raising tax capacity, tax capital at the moment..

I think your question around our guidance, and I think Alex mentioned this in the prepared remarks in the script, our guidance assumes there is no change. It's just too speculative at this point in time to make any significant assumptions as what could happen.

Way too many moving pieces, so we're assuming that effectively all the components whether it's the ITC, whether it's the depreciation, interest expense deductibility, we're assuming all that stays as is..

And our next question will come from Philip Shen with ROTH Capital Partners. And our last call will come from Philip Shen. Go ahead, please. Again, Philip Shen, your line is open, please go ahead with your question. Once again, Philip Shen, your line is open. Please proceed with your question. Sure, sure, we will.

Our next question will come from Colin Rusch with Oppenheimer..

Thanks so much.

Guys, as we think about the cadence of bookings as we go through the balance of 2017 into 2018, when do you start really focusing on the Series 6 product and building a backlog for that besides your own project backlog?.

Yeah. So, I mean, it's a good question. It's one of the things that we highlighted. So, today, we spent a lot of time talking through the product readiness and what we're doing in that regard.

The next call, Q1, we'll talk more from a market readiness standpoint, and there is a lot that needs to be done in that regard, right, in terms of understanding the spec, developing what we refer to as a PAN file that ultimately is used to do the energy prediction.

There's engagement with independent engineers, right, that they have to be involved with as well, so they can help provide kind of that third-party voice to our customers, and as it relates to the product and its performance and quality standards and everything else. So, there is a lot that we need to do.

What I will tell you is that there's been a tremendous response so far from our customers. Actually, I was talking with one of them yesterday as well and everyone wants to be a launch for Series 6. Now, the way we've described it to a lot of our customers is, the vast majority.

And if you look at our, what we'll report in our 10-K, which will be filed tonight and be available tomorrow, we're going to show close to 2 gigawatts of projects in our contracted pipeline.

Now, some of that is near-term, some of that will be Series 4, but really, easily can look at that pipeline and there's going to be north of 1-gigawatt on a DC basis of opportunities for Series 6, given the timeline of those projects. The second half of 2018, we'll be producing 1-gigawatt of Series 6.

So, the early production will go to our own projects. So, we're really talking about having the opportunity to sell through Series 6 starting in the 2019 timeframe. Now, module-only activity in bids aren't really happening today per se out in that horizon. Some markets, yes; for the vast majority, no.

Development opportunities will start to happen out in that horizon. So, we can use that Series 6 and the leverage of Series 6 to bid into development assets as well in a longer dated horizon. So, we do expect and we do have pipeline.

We highlighted in today's call, we've got at least in mid-to-late stage negotiations about 500 megawatts right now of Series 6. That all should drive momentum. Hopefully, we'll realize in some of those bookings as we progress throughout the year, but the pipeline will continue to build.

We'll start to see bookings for Series 6 in the second half of the year, but we clearly would expect a much stronger activity around confirmed bookings for Series 6 as we get into 2018..

Thank you. And that does conclude today's conference call. We do thank you for your participation today. Have a wonderful day..