Fluent Essays

Attached, please find the case study to be used for your final exam. Please read the case study carefully and answer the following questions listed below. Please use about 1 page to answer each question. Please return the final exam to me by 6pm on Saturday, May 2. - Please explain the existing and growing marketplace for the PV sector, including the dominant markets of producers and users. - Describe briefly the history of First Solar. And what were the opportunities and challenges for First Solar? - Understanding the situation First Solar finds itself in, what do you believe would be the best option for the company, in particular in its international business activities? - How can this case study be applied to the topics covered during class (International Business)? - What did you learn from this case study that could be applicable to other instances of international business? This case was prepared by Jennifer Ballen, MBA 2017, and Professor Neil Thompson. Copyright © 2017, Neil Thompson and Jennifer Ballen. This work is licensed under the Creative Commons Attribution- Noncommercial-No Derivative Works 3.0 Unported License. To view a copy of this license visit http://creativecommons.org/licenses/by-nc-nd/3.0/ or send a letter to Creative Commons, 171 Second Street, Suite 300, San Francisco, California, 94105, USA. 17-181 September 13, 2017 First Solar Neil Thompson and Jennifer Ballen Tymen deJong, First Solar’s senior vice president of module manufacturing,1 fixated yet again on the company’s latest 10-K. DeJong had joined the company in January of 2010, at a time when First Solar’s future appeared bright. Now, just two years later, First Solar’s cost advantage was eroding and deJong was facing challenges that would require tough decisions. In 2009, First Solar broke cost records by becoming the first photovoltaic (PV) manufacturer to produce panels that generated a megawatt of power at a manufacturing cost of less than $1.00 per watt.2 The company’s proprietary thin-film cadmium telluride technology had made it the largest and lowest-cost producer for nearly a decade. However, the 2011 Form 10-K on deJong’s desk revealed a net operating loss of $39 million, the company’s first year-end net operating loss in the past seven years. Although revenues were $2.7 billion, revenue growth had slowed from 66% in FY 2009, to 24% in FY 2010, and then to a meager 8% in FY 2011.3 Much of this slowed growth was attributable to broader trends affecting the entire PV industry. Chinese manufacturers, subsidized by their government, were flooding the market with low-price crystalline-silicon (c-Si) solar panels. Market demand for PV panels was also weakening. The 2008–2009 global financial crisis had squeezed government budgets and weakened the financial positions of many banks. As a result, the once-heavy European solar subsidies were shrinking and the willingness of banks to finance solar projects had virtually disappeared. Silicon raw material 1 As of July 2015, Tymen deJong became the chief operating officer (COO) of First Solar. 2 Watt: a unit of power is defined as 1 joule per second; it measures the rate of energy flow. 3 First Solar Inc., Form 10 K, 2007. FIRST SOLAR Neil Thompson and Jennifer Ballen September 13, 2017 2 prices were also falling. This helped First Solar’s competitors, which produced silicon-based panels, but not First Solar, which produced cadmium telluride-based ones. As deJong reflected on the company’s recent financial slump, he wondered if First Solar’s competitive edge had eroded permanently. How should First Solar respond to the threat from the Chinese manufacturers? What could the company do to maintain its cost advantage? Were First Solar’s recent acquisitions of down-stream solar panel installers a strategic benefit or a distraction? DeJong knew that to answer these questions, he first needed to better understand the sources of First Solar’s competitive advantage and whether these sources were sustainable. PV Solar Manufacturing and Distribution Solar Industry History and Evolution In 1839, nineteen-year old French scientist Edmond Becquerel discovered the photovoltaic effect: that shining light on the junction of two dissimilar materials, such as a metal and a semiconductor, creates electric current. This led to Bell Lab’s 1954 creation of the first functional solar cell. Early solar cells were inefficient and costly to manufacture, so their use was limited to high-value applications, such as space satellites.4 By the early 1980s, PV solar cell use had broadened to consumer applications, such as calculators and watches, and by the mid-1990s utility companies had begun using PV solar plants, although costs continued to be higher than nonrenewable energy sources. At the turn of the 21st century, two major types of solar technologies had emerged: solar thermal and photovoltaic. Solar thermal power plants used sunlight to generate heat that was used to boil water, with the resulting steam driving a turbine to create electricity. But, the fastest growing solar market was photovoltaics: the conversion of sunlight directly into electricity. First Solar produced exclusively photovoltaic panels Overview of Photovoltaics By early 2012, there were two dominant technologies used to produce PV solar power: (i) thin-film and (ii) crystalline silicon (c-Si) (Exhibit 1). The PV supply chains typically involved the following steps (Figure 1). 4 “Solar Explained: Photovoltaics and Electricity,” U.S. Energy Information Administration, October 25, 2015. FIRST SOLAR Neil Thompson and Jennifer Ballen September 13, 2017 3 Figure 1 Steps in the PV Supply Chains Production Stage Process for Crystalline Silicon Process For Thin Film i) Raw material preparation Raw silica, often in the form of sand, is purchased and purified. A substrate (e.g. glass) and semiconductor (e.g. cadmium telluride, CdTe) are prepared by 3rd parties. ii) Solar wafer production Silicon is formed into thin circular wafers. N/A iii) Solar cell production5 Solar wafers are layered to generate electric current when hit by sunlight. A thin layer of semiconductor is layered on top of the substrate, coated, and then defined with a laser. iv) Module array production Solar cells are electrically wired together into solar modules and weatherproofed. v) System integration and development System integrators install completed modules and arrays. For utility customers, integrators also provide financing, engineering, construction, and ongoing maintenance. Source: Case writers. Crystalline silicon was the dominant technology in the market, accounting for nearly 85% of manufactured solar panels over the last decade. Crystalline silicon was used for semiconductors in both electronics and solar cells. In 2001, 20% of total silicon use was allocated towards solar cell production, and 80% towards electronics. By 2010, this had reversed: 80% of total silicon use was for the manufacturing of solar cells. The rapid growth in demand from solar manufacturers increased silicon prices from $50/kg in 2001 to a peak of $475/kg in 2008.6 In response, crystalline silicon manufacturers raced to improve cell efficiency and reduce the thickness of the silicon wafer, which decreased silicon use in solar cells from approximately 15 grams per watt in 2001 to 5 grams per watt by EOY 2011.7 From 2008–2011, supply of silicon ramped up, causing prices to plunge from $475/kg back to $65/kg (Exhibit 2). Industry experts predicted that silicon prices would continue to decline further in the near future, benefiting First Solar’s competitors. An alternative to crystalline silicon was thin film technology, first commercialized in the early 2000s by First Solar and a small number of other manufacturers. True to its name, thin film technology involved the placement of thin layers of semiconductor material, such as cadmium telluride, on top of inexpensive substrates, such as glass or aluminum. Panels using thin film were typically lower cost and required 98% less semiconductor material than traditional c-Si panels. In 2011, cadmium telluride use in thin film solar panels was approximately 0.1 grams per watt. The price of cadmium telluride varied 5 “The Difference Between Solar Cells and Solar Panels,” RGSEnergy.com. 6 “Mineral Commodity Summaries,” U.S. Geological Survey, January 2012. 7 Shyam Mehta, “The Shifting Relationship Between Solar and Silicon in Charts,” Greentech Media, 2011. FIRST SOLAR Neil Thompson and Jennifer Ballen September 13, 2017 4 over time, from $48/kg in 2006 to $192/kg in 2011 (Exhibit 2). Offsetting thin-film’s cost advantage was its historically lower efficiency in converting sunlight into power for most applications (Exhibit 3). The cost of nonrenewable fossil fuel power had historically been lower than that of renewable power. By the end of 2010, ignoring subsidies, it cost utilities approximately $0.15-$0.35/kWh to produce electricity from solar power, $0.08-$0.10/kWh to generate electricity from wind, and $0.06-$0.08/kWh for natural gas.8 Coal cost only $0.04/kWh, but was the dirtiest form of power. Indeed, many coal plants with remaining useful life were being decommissioned to avoid the environmental and health damage they caused. Natural gas was becoming cost-competitive with coal due to the reduced cost of extracting natural gas through hydraulic fracking,9 a technique that had increased in use substantially over the past decade. However, natural gas, while cleaner than coal, still produced carbon emissions and posed environmental risks. Historically, the cost of solar was much higher than other forms of power. In 1976, the cost of solar was approximately $2.00/kWh, but this cost was falling substantially as producers learned-by-doing and took advantage of economies of scale (Exhibit 4). Global Market Over the last decade, PV solar energy had become the fastest-growing power generation technology in the world. Much of this growth was driven by regulatory policies, as solar was still more expensive than traditional fossil fuels. Government incentives typically enhanced the returns for solar providers in two ways: either providing higher prices for solar power suppliers or requiring utilities to purchase a specific amount of solar power.10 For example, Feed-in Tariffs (FiTs) were widely used, particularly in Europe, and offered solar producers long-term contracts at above-market, government-mandated rates. Another incentive, termed renewable portfolio standards, mandated that certain percentages of the energy produced by utilities be sourced from renewables, such as solar, wind, geothermal, or hydroelectric power. Renewable portfolio standards were used by many states in the United States, most significantly California that had been increasing renewable percentage requirements since 2002. From 2002–2008, global PV demand increased at an average annual rate of 48%. However, in early 2009 the global financial crisis impacted the solar market, tightening the wallets of financial institutions and decreasing government spending. Existing subsidies allowed demand to continue increasing, but at a slower rate, after 2009. By early 2012, many governments had significantly reduced incentive programs. This was particularly evident in Europe, whose share of overall demand fell, albeit from a 8 “Electricity Generation Estimates,” U.S. Energy Information Administration and Michigan State University, April 2011. 9 Hydraulic fracking is an extraction technique for oil and gas wells in which pressurized liquid is injected into the cracks in rock formations. Once the hydraulic pressure is removed from the well, the remnants of the fracking fluid ease the extraction of oil and gas. 10 Government incentives came in many different forms including, but not exclusive to: feed-in-tariffs, renewable portfolio standards, quotas, tax credits, tendering systems, net metering, rebates, loans, and production incentives. FIRST SOLAR Neil Thompson and Jennifer Ballen September 13, 2017 5 high level (Exhibit 5). Despite this, the total global PV installed base at EOY 2011 was 65 gigawatts and experts predicted that this would grow by 400-600 gigawatts by 2020.11 The biggest change in solar production was the large-scale entry of Chinese producers. In 2001, China comprised less than 1% of overall solar production, but by 2012 Chinese producers were manufacturing nearly 60% of the entire world’s supply of PV panels12 (Exhibit 6). Market Segments There were three broad markets for solar power: residential homeowners, commercial businesses, and utilities. The residential segment represented 29% of the total market and was predicted to grow to 35% by 2020. Commercial businesses comprised 40% of the market; this segment was expected to shrink to 25% by 2020. The utility market was predicted to be the fastest growing segment, with an expected increase in market share from 31% in 2011 to 40% by 2020. In all three markets there were numerous systems integrators. The Residential Market In the residential market, PV solar manufacturers sold panels to third-party system integrators, installers, and distributors, who would physically position the panel on a homeowner’s roof and connect the panel to the regional electric grid. Residential users were encouraged to adopt solar through investment tax credits and net metering incentives (which encouraged solar operators to sell unused electricity back to utilities). Residential customers typically did not focus on the technology or maker of their solar panels, but instead on the overall costs and benefits of the installed system. The key criteria for a residential customer purchasing from a panel manufacturer were (in descending order): the levelized cost of electricity (an average cost measure per kWh across the lifetime of the system),13 installation and distribution costs (expenses that were paid by the homeowner), watts per unit area, and sometimes even aesthetics, as some residential homeowners were concerned about the appearance of highly visible rooftop panels. The Commercial Market Commercial and industrial businesses seeking to lower their operating expenses and carbon footprints also purchased solar power systems through third party system integrators and distributors. As commercial projects were typically larger in scope and required greater wattage per panel, the primary purchase consideration for commercial businesses was the levelized cost of electricity. When purchasing panels, commercial customers also focused on watts per unit area, installation and distribution costs, and reliability of the technology. 11 Krister Aanesen, Stefan Heck, and Dickon Pinner, “Solar Power: Darkest Before Dawn,” McKinsey & Company, May 2012. 12 Robert Castellano, “China’s EV Battery Industry Could Be A Repeat of Solar and Rare Earth Dominance,” Seeking Alpha, October 25, 2016. 13 See Glossary for more details. FIRST SOLAR Neil Thompson and Jennifer Ballen September 13, 2017 6 The commercial and utility markets both financed solar projects with solar leases and power purchase agreements (PPAs), financial contracts between buyers and providers of electricity. With a PPA, the developer was responsible for the design, financing, and installation of the solar system at little to no cost for the customer. The developer also operated and maintained the system over the duration of the contract, typically 10-25 years. In return, the customer purchased the generated energy at a fixed rate from the developer. At contract termination, the customer would either extend the PPA, remove the system, or purchase the system from the developer. PPAs provided an assurance of both volume (all the kWh were sold) and price (as set by the PPA contract). The Utility Market In contrast to the residential and commercial markets, the utility market encompassed a smaller number of larger-scale projects. For example, in the United States, there were approximately 60 new utility-scale solar projects in 2011, as compared to hundreds of thousands of projects in residential and commercial markets.14 Some utilities purchased panels directly from PV manufacturers, while others purchased from system integrators and installers. System developers provided a variety of services to utility customers, including: i. Project Development: obtaining land permits, negotiating purchase agreements, transmission interconnection, major engineering, and construction. ii. Operations and Maintenance: subsequent to development, signing long-term contracts to provide on-site operations and maintenance, such as performance analysis, forecasting, contractual and regulatory advice, performance reporting, and inventory management. iii. Project Finance: negotiating and executing power plant sales, raising capital from debt and equity markets, and structuring non-recourse project-level debt financing. iv. Engineering, Procurement and Construction: engineering and designing power plants, developing grid integration, construction management, and procuring component parts from third parties. The primary purchase consideration for the utility market depended on the placement. In space- constrained areas, the most important factor was typically watts-per-square meter, so that as much power as possible could be generated in small spaces. Utilities that were not space constrained were willing to purchase less efficient panels if the panels had a lower cost per kilowatt-hour. Many utility installations were not space constrained. A vendor track record of successful and timely installation was typically the next purchase consideration for utilities. PV manufacturers that wanted to sell products to utilities in a certain location would often first establish a relationship with integrators that had a favorable track record in order to better reach that market. Finally, utilities purchased panels based on proven technology and anticipated 14 “An Analysis of New Electric Generation Projects Constructed in 2011,” Electric Market Reform Initiative and American Public Power Association, March 2012. FIRST SOLAR Neil Thompson and Jennifer Ballen September 13, 2017 7 reliability of the system. Feed-in-tariffs were implemented by many governments to encourage demand and required utilities to buy renewable energy at above-market rates. Utilities often passed this incremental financial burden to their customers through a small extra fee on monthly electric bills. First Solar Brief Company History First Solar originated as a glass company in 1984 under the name Glasstech Solar, founded by glass entrepreneur Harold McMaster. In 1990, the company was renamed to Solar Cells, Inc., and then once again in 1999 to First Solar, LLC, after True North Partners purchased a controlling interest in the company and the firm was recapitalized. John Walton, the son of Walmart’s founder Sam Walton, and Mike Ahearn (who later became co-founder, Chairman, and the first CEO of First Solar) founded True North Partners. Walton and Ahearn both believed in the power of technology to accelerate sustainability. On November 17, 2006, First Solar became a publicly-traded company (FSLR), raising $450 million at an initial offering price of $20 per share.15 First Solar’s business model focused solely on component manufacturing at first: designing and producing PV solar cells and modules to sell to project developers, system integrators, and operators of clean energy projects. Beginning in 2007 with a series of acquisitions, First Solar vertically integrated, buying system integrators primarily in the United States. Through its systems business, First Solar controlled the engineering, procurement, construction, operations, maintenance, and development of solar power plants, and at times, project finance. Manufacturing and Costs First Solar manufactured PV solar cells and modules using an advanced thin-film cadmium telluride (CdTe) technology, controlling all stages of production entirely in-house which, according to First Solar’s 10-K, “…eliminated the multiple supply chain operators and expensive and time consuming batch processing steps that are used to produce crystalline silicon solar modules.” In 2005, First Solar produced its first commercial solar module. First Solar used a proprietary vapor deposition technology to coat glass panels with two thin layers of semiconductor material: first cadmium sulfide, then cadmium telluride. High speed lasers then divided the semiconductor into cells, the fundamental units for absorbing light and converting it into electricity. Solar cells were combined to form solar modules and solar modules were combined to form solar panels to scale up the amount of electricity provided. Tymen deJong commented on First Solar’s use of thin-film: 15 Nasdaq, First Solar Inc. IPO priced November 17, 2006. FIRST SOLAR Neil Thompson and Jennifer Ballen September 13, 2017 8 Most of the early work in photovoltaic generation was done on crystalline silicon, so that’s where the R&D investments went. While there was an awareness of thin-film and cadmium telluride, there simply was not that much money being invested in it. There are significant technical challenges in applying cadmium telluride. We figured it out early and, to this day, we have a tremendous amount of IP around how to do that. The barriers to entry to figure this all out are years of R&D and hundreds of millions of dollars in capital expenditures. And, to be fair, all of the early efficiency records were based on c-Si…it looked like a better technology to new entrants. But, if you want to look at thin-film, you have to do all that work yourself. Our company leaders had this vision around CdTe and what we could do. Historically, First Solar produced all of its modules at its manufacturing plant in Perrysburg, Ohio, which later evolved to also become the company’s primary research and development (R&D) center. In April of 2007, First Solar expanded production internationally and began to produce modules at its Frankfurt/Oder Germany plant. As of 2011, First Solar operated 36 production lines in Perrysburg, Ohio; Frankfurt, Germany; and, Kulim, Malaysia. Of these, the Malaysian plants had the lowest production costs, but the other plants had advantages in terms of R&D or serving particular markets. The company’s newest plant was built in Frankfurt, Germany in November of 2011. This was First Solar’s second plant in Frankfurt, adding a capacity of 250 megawatts per year to the region. The plant had taken First Solar one year to construct and cost roughly 170 million euros (US $230 million).16 First Solar also had two plants under construction in Mesa, Arizona and Ho Chi Minh City, Vietnam.17 Traditionally, First Solar had operated its plants very close to 100% capacity in order to maximize use of the expensive fixed capital required to produce PV panels. By 2011, however, the increasing market share of Chinese competitors led to First Solar producing only 1.7 gigawatts of panels (approximately 21 million solar modules) despite having the capacity to produce 2.5 gigawatts. The manufacturing cost per watt for First Solar and its competitors is shown in Exhibit 7. Customer and Market Strategy The majority of First Solar’s early customers were system integrators, developers, and operators, primarily located in subsidy-rich Europe. In 2008, approximately 74% of the company’s net sales resulted from Germany alone.18 In order to diversify, First Solar expanded into direct sales in high- sunshine, non-subsidy reliant markets, primarily selling systems to utilities in Africa, the Middle East, and the Americas. 16 Jonathan Gifford, “First Solar Inaugurates Second German Plant,” PV Magazine, November 3, 2011. 17 First Solar Inc., Form 10-K, 2011. 18 First Solar Inc., Form 10-K, 2010. FIRST SOLAR Neil Thompson and Jennifer Ballen September 13, 2017 9 The company first ventured into the systems business in late 2007 with a $34.4 million acquisition of system integrator, Turner Renewable Energy. Further acquisitions of Mission Edison’s project pipeline, OptiSolar (a power plant contractor), NextLight Renewable Power (a solar panel developer), and Ray Tracker (a component parts firm), expanded First Solar’s presence in the systems market.19 While First Solar became closer to the customer, these acquisitions also brought with them higher SG&A expenses. From 2009 to 2011, First Solar grew its utility-scale systems business from 5% to 25% of overall sales, narrowing the gap between itself and systems leader, SunPower, which derived 53% of its business from systems in 2010 and 46% in 2011. Chinese manufacturers were largely absent from the systems business. Exhibit 8 provides additional details. Financial Strategy First Solar pursued a conservative financial strategy, borrowing less than its competitors. From 2007– 2011, First Solar had an average annual debt of $276 million, whereas SunPower had $687 million, Suntech had $1.7 billion, and Yingli Solar had $1.1 billion. First Solar also consistently kept more cash on hand than competitors, for use in financing promising solar projects. Capacity expansions were typically funded with 50% cash and 50% equity. Bruce Sohn, former President and Board member (2003–2011), commented on First Solar’s financial approach: The reason we pursued a low leverage strategy was because we wanted a strong balance sheet. This served to both lower borrowing costs [for First Solar customers] and provide confidence to buyers that we would be able to sustain our business for the long-term. We did it by design for those reasons. In contrast, our competitors during this time were levering up and borrowing to expand, and thus had weak balance sheets. People didn’t trust those companies. First Solar took the opposite approach. Exhibit 9 shows both the income statements and balance sheets for First Solar and its main competitors. Vertical Integration All PV manufacturers produced solar modules, with several outsourcing various aspects of semiconductor production. Few forward-integrated into systems, so First Solar was unusual in this respect. The company divided its business into two interrelated segments: components and systems. The components business manufactured cadmium telluride solar cells and modules, while the systems business developed those components into complete solar systems. The components segment had historically achieved higher profitability and generated more cash than systems, but the systems business had less margin variability because the provision of ongoing maintenance, engineering, and construction was less dependent on materials prices. 19 First Solar Inc., Form 10 K, 2007. FIRST SOLAR Neil Thompson and Jennifer Ballen September 13, 2017 10 Sohn commented on the vertical integration: We realized that we could scale our production faster than our customers [the systems integrators] were able to scale their business. Our customers were the constraint and we determined that if we could vertically integrate, especially in places where our customers did not operate, then we could grow significantly faster. This effectively doubled our shipment rate and enabled steep volume growth even during a period of heightened competition. Having our own utility scale solar business also provided us with the opportunity to optimize overall system design…For several years, First Solar was able to deliver systems that yielded up to 5% better performance than competitors because of our intimate knowledge about the [First Solar] panels. Competition United States Although U.S. customers were initially slower to adopt PV solar power than their European and Asian counterparts, by 2011 U.S. solar installations had grown enormously, doubling from 2009 to 2011. In 2011, the U.S. market share of total global PV installations increased from 5% to 7%. U.S. market share was anticipated to outpace the growth of other nations over the next five years. Reported installed solar capacity from 2010–2011 in the United States was a total of 1,855 megawatts, comprised of 16% residential, 43% commercial, and 41% utility. The utility market had only recently grown in size, while the commercial market had long accounted for over 50% of solar energy growth.20 As in the rest of the world, the majority of modules produced in the United States used crystalline silicon technology. In 2011, First Solar controlled approximately 41% of the U.S. market. SunPower was the second largest PV manufacturer, controlling 38.5%, while the remaining 20.5% of the market went to smaller players including Solyndra, SunEdison, SunRun, Evergreen Solar, and Spectrawatt, Inc.21 SunPower manufactured highly efficient (18.1%–20.1%) and more expensive, solar panels and modules. In 2011, SunPower was suffering a similar fate to First Solar, also recording its first year-end net operating loss since 2007. SunPower’s gross margin over the past five years had decreased from 19% in 2007 to 10% at EOY 2011. In April 2011, SunPower sold a 60% controlling interest to the oil company Total for $1.38 billion. Total offered SunPower up to $1 billion of credit over the ensuing five years.22 Solyndra, a California-based solar panel manufacturer, also competed in the thin-film …