To the New World

Sure, SunPower Corp., which makes the most efficient commercially available solar panels in the world, is based in Silicon Valley. But the company has always made its panels elsewhere. It built its first factory in the Philippines in 2006, and also manufactures in China and Mexico. Now, for the first time, the company plans to set up a factory in the United States, in its home state of California.
SunPower announced in April that it would partner with Singapore-based contract manufacturer Flextronics to begin making solar panels in Milpitas, California, by the end of this year. The project is expected to have an initial capacity to produce up to 75 megawatts worth of panels annually. “We’ve said for years we planned to put manufacturing close to end markets and this is proof,” said Vice President Julie Blunden. “The U.S. market is one of the fastest-growing markets in the world today, and if you look at the total available market in the long run, it’s just a very large power market that’s got a lot of room to grow.”
The homecoming represents one of dozens of new solar manufacturing facilities cropping up in North America, a market which has long held the potential – but not the policies – to reach industry-moving size. Among many examples, Atlanta, Georgia-based Suniva, which already produces screen-printed monocrystalline cells at its 100-megawatt factory in Norcross, Georgia, last year announced it would expand its existing plant’s capacity by 75 percent and build a new 400-megawatt factory in Michigan that will begin operating – with a smaller initial capacity – next year. In the U.S., at least 11 new manufacturing plants are expected to be built by 2012, a big jump from the 27 solar factories in the country at the beginning of 2009, according to Shyam Mehta, a senior analyst at Greentech Media.
That number only includes plants that already have scored financing, so the real number could be larger. And there are several indications that the trend will skyrocket beyond 2012. For one thing, take U.S. federal manufacturing tax credits, which materialized as part of the stimulus bill. So far, 61 new or upgraded manufacturing facilities – expected to cost a total of 3.8 billion U.S. dollars – have been awarded U.S. federal manufacturing tax credits worth 1.56 billion dollars, according to the Solar Energy Industries Association (SEIA). In addition, project announcements have been accelerating, Mehta says. More plants were announced in the first half of last year than in the previous three years combined, he wrote in a report. All together, by the end of 2012, Mehta expects U.S. cell capacity to reach four gigawatts, up from 786 megawatts in 2008, and panel capacity to total 3.88 gigawatts, up from 875 megawatts in 2008. That 2012 capacity would enable the production of 2.79 gigawatts of cells and 2.69 gigawatts of panels per year.
Canada also is expecting to see solar manufacturing grow in the next few years. Canadian Solar, which currently manufactures crystalline ingots, wafers, cells and panels in China, is now building its first North American facility in Ontario, Canada. The 200-megawatt panel factory is expected to begin operations – with a smaller initial capacity – this year. And U.S. microinverter company Enphase Energy, based in Petaluma, California, in March announced it would build a 100-megawatt production line in Ontario.
It’s not just North American companies that are moving manufacturing to the continent, either. European and Asian companies also are building factories there. China’s Suntech Power Holdings, for example, in January unveiled plans to construct a solar-panel manufacturing plant in Goodyear, Arizona, about 30 kilometers west of Phoenix. The factory, expected to begin with 30 megawatts of initial capacity, could grow to more than 120 megawatts, according to the company. Suntech, the world’s largest crystalline-silicon solar-panel maker, said it’s the first Chinese cleantech company to bring manufacturing to the U.S.
Meanwhile, Germany’s SolarWorld said in May that it’s nearly finished expanding its Oregon solar-cell factory, which has 150 megawatts of capacity at the beginning of last year, to a whopping 500 megawatts. And in Athens, Greece-based HelioSphera, an Oerlikon micromorph customer, which in September launched production at its first factory, a 60-megawatt thin film panel facility in Tripolis, Greece, in December announced it would begin building a 160-megawatt facility in Philadelphia, Pennsylvania, this year. The company, previously called Next Solar, expects to reach full production at its Pennsylvania plant in 2012.

Manufacturing magnet

So why is this shift happening now, amid a recession rife with financing difficulties? The main reason is policy change. The U.S. added a raft of new incentives – including grants, loans, tax breaks, rebates and contracts – for clean energy with its federal stimulus bill last year and bailout bill in 2008. For manufacturing, one of the most significant stimulus provisions was the 2.3 billion dollar Advanced Energy Manufacturing Tax Credit, which gives clean-energy companies a tax break on 30 percent of the capital cost of manufacturing equipment. Canada also approved increased grants, research funding and other investments for green energy in its stimulus plan, although the spending amounted to 18 times less, per person, than the U.S. stimulus plan, according to a study by Environmental Defence.
Important incentives are coming not just from the federal government, but also from local governments, including individual Canadian provinces – most notably Ontario, which passed a feed-in tariff paying more for projects with a certain amount of domestically produced content – and U.S. states, which are trying to meet renewable portfolio standards that require utilities to get a certain percentage of their electricity from renewable sources. HelioSphera, for example, is getting 49 million dollars in grants and loans from Pennsylvania to help defray the anticipated 500 million dollars cost of its factory. And Suniva is getting a 15 million dollar tax credit from Michigan, on top of a 141 million dollar federal loan guarantee, for its 250 million dollar factory in Michigan. All together, 18 states offer their own incentives for solar manufacturing.
For SunPower, the decision to manufacture in California involved both policy changes and technology advances. First of all, the company received approval for 10.8 million dollars in federal manufacturing tax credits and 24 million dollars in funding for research and development from the U.S. Department of Energy’s Solar America initiative. SunPower also scored a 1.5 million dollar loan from the city of Milpitas, and is eligible to take advantage of a variety of California grants, loans and bonds, as well as funding for employee training. In addition, state policies helped provide a clear view into the market, Blunden said.
Aside from state incentives and rebates for solar projects, a new California rule, signed into law in March, offers a sales tax break for green-manufacturing equipment, and a tax bill ensured that federal Treasury grants defraying the cost of renewable energy projects wouldn’t be taxed by the state. The state also doubled the net metering cap, which previously limited the amount of solar power utilities could credit back to customers at 2.5 percent, to five percent. “It’s really interesting to see how things have worked out this year,” Blunden said.
Meanwhile, research conducted under the Solar America initiative yielded a design for a more automated production line, which eliminates workers from some steps – such as transitions between different tool sets – that previously required people. The line, which reduces the amount of labor needed to produce solar panels, will enable SunPower to competitively manufacture in areas with high labor costs, such as in North America and Europe, Blunden said.
While analysts such as Mehta say the cost of financing manufacturing plants in China is often still cheaper than in the U.S., especially for Chinese manufacturers, SunPower says the cost of its U.S. plant will be similar to that of its Chinese manufacturing, all things considered. “It’s comparable because we have lower logistics costs, and it is expensive to put relatively heavy solar panels on the water and transport them as inventory for multiple weeks,” she said.
In other words, SunPower is confident it will be able to sell all of its panels from the California factory in the U.S. and will reduce transportation costs by manufacturing in the country. Not only is the U.S. one of the fastest-growing solar markets in the world today, but it’s also a very large power market, meaning that its solar market’s got plenty of potential room to grow, Blunden said. The country could be the world’s largest market in the next five years, she added.
And that’s been a huge driver of North American manufacturing: the conviction that the U.S. and Canada will soon become big solar markets, and that building North American factories will provide an advantage in targeting those markets. As the solar market in Germany starts to shrink as a result of a lower feed-in tariff, and analysts start to wonder if some European markets are becoming saturated, many have predicted the U.S. could be the next big solar market in the next few years.
Because solar panels are made of heavy glass, it’s expensive to transport them, which is why SunPower and other manufacturers want to produce panels near their end markets. But the financial markets have helped out there. The dollar is worth less than the euro, with one euro equaling 1.25 U.S. dollars in May, and that’s allowed manufacturers to build factories in the States and sell profitably into the European market, in spite of the extra transportation costs, said Rhone Resch, SEIA’s Chief Executive. Meanwhile, planting manufacturing capacity in the U.S. allows companies to stake a position in case the North American market takes off, he adds. However, that dynamic could be changing as the euro’s value falls, dropping 18 percent against the dollar since November.
The U.S. has proven particularly attractive to thin film manufacturers, and Mehta expects thin film manufacturing to continue to make up most of the U.S. production in the next two years.

Top U.S. product? Thin films

Mehta predicts that thin films will make up 2.69 gigawatts, or 67 percent, of cell capacity in the country, enough to make 1.82 gigawatts of cells by 2012. The reasons? Much of the thin film research is coming out of U.S. labs right now, and China’s lack of intellectual property protection dissuades many companies with new technologies from setting up there, Mehta said. “Thin film has got a lot more IP, so companies are going to be wary of constructing a fab in China, because they’re afraid their IP will get stolen,” he said.
Still, he expects both thin film and crystalline manufacturing to grow, with most of that growth happening at the panel level. In Mehta’s report, he wrote he is seeing “a marked preference for building module-assembly plants compared to wafer and cell facilities, which reflects the U.S.’s status as an anticipated leading end market, as opposed to a low-cost production location.” Even if most of the components are made outside of the States, making the panels in the country wins companies political points, he said.
That said, Mehta sees plenty of opportunity for feedstocks, glass, encapsulant and equipment suppliers to supply those module assemblies. The U.S. has long been a key polysilicon provider – supplying 40 percent of the world’s polysilicon – because of its background in semiconductors, Resch said. Mehta said he’s seeing less U.S. activity in components such as inverters. Ontario, with its feed-in tariff based on domestic content, has attracted inverter companies. Aside from Enphase, which inaugurated its Ontario line in March, saying it would enable its customers to take advantage of the tariff, Boston-based Satcon said in March it is manufacturing its utility-scale inverters in Burlington, Ontario.
So far, most of the North American manufacturing projects haven’t been large. Essentially, companies are dipping a toe in the water to test the market temperature. Exceptions exist, such as SolarWorld’s 500-megawatt plant, but most companies are taking baby steps, Mehta said. After all, the U.S. market has disappointed solar advocates before, and it’s possible it won’t accelerate as quickly as many hope. Mehta expects the U.S. solar industry will continue to export more polysilicon and panels than it imports until at least 2013, and – especially if the currency exchange stops working in U.S. manufacturers’ favor – manufacturers face the risk of building too much capacity too soon and ending up at a disadvantage. Balancing production with the demand is an ever-present challenge, as the silicon and panel shortage, then the oversupply, in the last years testifies.
But Resch, who believes the U.S. is a net importer of panels, disagrees that the it will be exporting panels for the next years. “I think the U.S. market will be growing faster than what our module manufacturing capacity will be,” he said. “Unless manufacturing can keep up with the 40 percent growth rate we expect to see in the future, we’re going to increase our imports.”
In the past, manufacturing shifts – such as when manufacturing expanded from Japan to Germany and then to China – have always brought about greater impacts on the industry. If more manufacturing shifts to North America, what changes will that bring? Well, not as much as you might think, according to Mehta. Because most of the U.S. manufacturing is panel assemblies, much of the production will remain in other countries, such as Germany and China, with just that last step being taken in the U.S., he said. And many of the manufacturers operating in the U.S. will be international companies from Germany and China, he added. “It’s less than meets the eye,” he said.

Local value chain

Meanwhile, the Ontario feed-in tariff could end up raising prices. First of all, limiting the number of suppliers that can sell into the Ontario market will likely encourage less competition, said Adam Browning, Executive Director of solar advocacy group Vote Solar. “The chief barrier to establishing solar markets is cost, and these local content requirements frankly drive up the price by limiting the amount of potential suppliers.” At 80.2 Canadian cents per kilowatt hour, the region is clearly paying more to get locally derived products, but while that’s definitely brought in manufacturers, it runs the risk of building a local value chain that only can work in that local market, he added. In other words, the manufacturers that can be successful in Ontario might not be competitive anywhere else after the feed-in tariff ends.
As the overall solar market grows, Mehta expects North America, which accounted for nine percent of the wafer, cell and module production in 2009, to make up only seven percent in 2012, but also expects Europe, which made 20 percent of that equipment in 2009, to also contribute only seven percent in 2012. As PV becomes commoditized, manufacturing will ultimately end up in the lowest-cost locations in the end, he believes, although the location of the market will have some sway. “If the U.S. became a major market, the war for manufacturing between the U.S. and Europe would suddenly be weighted heavily to the U.S., that’s definitely true,” he said. “The manufacturing landscape will receive a big boost and it will make the U.S. the new Germany. What will that mean? Well, 50 percent of the manufacturing will still be in China.”

Entering solar manufacturing

One of the biggest impacts to come from increased North American manufacturing could be somewhat indirect. Resch is seeing a number of companies from other industries, such as the semiconductor industry, automotive industry, glass and electronics industries, entering solar manufacturing in the U.S. This kind of cross pollination could end up leading to innovations and developments still unfathomable today. “That’s the game changer, when you have companies like Dow Chemical and DuPont participate in the industry at this scale,” he said. “There’s almost no doubt when you bring that high-quality manufacturing with that kind of investing, you will achieve higher scale and efficiency that will result in lower cost of solar.”