By Harro Ten Wolde and Christoph Steitz
FRANKFURT (Reuters) - Solar companies could be well advised to look to chip makers to figure out how to survive a sector shake-out, as the consolidation of the semiconductor industry suggests it is all about either being big and basic or small and selective.
Following insolvency filings in the past year that have included six heavyweight solar companies in Germany and the United States, industry experts and analysts say those players who can boast neither scale nor specialization look doomed.
Like computers, memory cards and digital cameras, solar components such as cells - used in the panels installed on roofs to convert sunlight into electricity - are becoming increasingly commoditized, so companies need to sell them in large volumes to make money.
More specialized products such as inverters, which feed the electricity into a power grid, and polysilicon, a raw material for the cells that is difficult to produce, still offer attractive returns.
"There might be between 10 and 30 companies globally in the next 10 years, which will control the mass market," said Stefan Raithel, managing director of the European solar arm of SEMI, the global chip industry association, compared with hundreds making panels and cells now.
"Also, there is an industry for smaller players, working for the niche markets, and making more customized products," he said.
Both solar and chip sectors have suffered massive price declines and cheap Asian competition in highly cyclical but growing businesses. Q-Cells, once the world's biggest maker of solar cells, and from the chip sector, Elpida, the world's No.3 maker of dynamic random access memory (DRAM), are notable casualties.
Over the last five years, revenues in the global solar energy industry have increased more than five-fold, reaching $93 billion in 2011, compared with $17 billion in 2007, according to consultancy Solarbuzz.
Moore's law, named after the co-founder of chip maker Intel Gordon Moore, stipulates an industry trend of doubling the number of transistors that can fit on a chip every two years at about half the price, improving processing speed, computer memory capacity or the number and size of pixels in a digital camera.
Something similar is happening in the solar sector.
"Solar is not the best industry to be in (at the moment) because you have to compete on price," said Xavier Chollet, a renewable energy fund manager at Pictet who also covered the chip industry as an analyst for several years.
Costs are currently the big issue, while capacity will be next - thinner solar cells integrated in modules are expected to become mass products in about five years' time.
"The cells business is all about economies of scale," said HSBC analyst Christian Rath, who covers solar as well as semiconductor stocks. "Here, the Asian players have the advantage of simply being big, resulting in lower costs."
In addition, European governments are slashing subsidies for the solar industry, making it even harder for companies to compete.
Those who cannot keep up with big players may be better off going for niche markets with higher margins, as demonstrated by chipmaker Qualcomm which has become very successful by specializing in chips for cell phones.
"It is just like Picasso would do a better painting than someone who just came out of art school," Pictet's Chollet said.
SEMI's Raithel said companies such as SolarWatt and Sunways had very realistic chances of staying in the market.
SolarWatt makes monocrystalline and polycrystalline solar modules as well as turnkey solar power plants, while Sunways offers all components required for high-yield photovoltaic power generation, prompting Chinese peer LDK Solar to buy a majority stake in the company.
Analysts also expect U.S.-based SunPower to survive due to its very efficient modules as well as the backing of French oil group Total, which, too, owns a majority stake in the group.
BIG IS BEAUTIFUL, NICHE IS NICE
Solar companies could attempt to divest or exit volatile businesses, taking a leaf out of German chipmaker Infineon's book. It spun off parts of its loss-making memory unit Qimonda, at the time the world's second-largest DRAM company, in 2006.
In 2010 it also reduced its exposure to the volatile mobile phone market by selling its wireless chip unit to Intel and now focuses on the automotive sector and chips used by industrial clients to make their business more energy efficient.
Solar companies with a technological edge and a broad portfolio - such as Wacker Chemie - could choose a similar path, Pictet's Chollet said.
Wacker Chemie, the world's No.2 maker of polysilicon, a key ingredient used for solar cells, may want cut its exposure to the volatile solar sector.
Spinning off its solar business, which accounted for more than two thirds of the group's core profit in 2011, would still leave the company with its other businesses that supply industries such as semiconductors, chemicals and pharmaceuticals.
For Germany's Sovello, this move came too late.
The group - once a joint venture of Q-Cells, Evergreen Solar and Norway's Renewable Energy Corp - aimed to float on the stock market in late 2008 or 2009 but later scrapped those plans. Earlier this month, the group focused on solar cells and modules, filed for insolvency.
(Additional reporting by Matt Daily in New York and Noel Randewich in San Francisco; Editing by Erica Billingham)