Evergreen Solar was a darling of Massachusetts politicians until it fell apart. But there's plenty to learn from what went wrong at this once-promising company.
If there's a lesson to be had in the wrenching downfall of Evergreen Solar, it's this: Even Apple doesn't try to do everything itself.
For two days now, the green-tech industry--and more than a few politicians who had put their credibility on the line to support it--have been wringing their hands over Evergreen, which on Monday announced it filed for Chapter 11 bankruptcy and will seek to sell off its assets to pay off hundreds of millions of dollars to creditors. Today Evergreen said that Nasdaq informed the company that its stock will be delisted next week.
Evergreen, whose stock has fallen from more than $100 three year ago to less than 20 cents now, has become a black eye for state officials who gave the company more than $50 million in aid to build a factory at an abandoned army base in Devens, Mass. (For a financial post-mortem, see SmartPlanet's look at its books.)
So what happened? The reality of global economics, it seems, got in the way. Low-cost manufacturing has slashed prices so fast that some suppliers simply can't make product at a profit. While that doesn't mean the solar manufacturing industry in this country is a goner, it does mean that companies are going to have to move toward more advanced technologies--and find the financial backers willing to stick it out for the long haul.
"If you can produce a more efficient, cheaper solar cell, the opportunities are enormous--if the technology is superior, you can manufacture it anywhere," said Ethan Zindler, head of policy analysis at research company Bloomberg New Energy Finance. "(But) the moment a product becomes commoditized, it becomes very difficult for the U.S. to compete."
• Why solar start-ups need Uncle Sam
• Falling solar costs: Good for buyers, bad for makers
• Evergreen Solar files for bankruptcy
• SmartPlanet: Sign of bad decisions, not industry's demise
Just like components in a PC or cell phone, solar cells and panels have become, for the most part, commodities. Chinese manufacturers that make solar panels with traditional polycrystalline silicon cells now hold 50 percent market share; other manufacturers have moved module, or panel, production to other places in Southeast Asia.
Look at it this way: 30 years ago, IBM created the modern model for PC manufacturing. The PC maker became a master assembler and let other companies build the shell, the chip, even the motherboard. It was a dramatic new way of doing things. While Apple today is considered the pre-eminent example of a tech company that does everything itself, it doesn't really. Apple comes up with the big ideas, the innovations, and lets someone who owns a massive factory in China, like Foxconn, actually do the building.
The commodity manufacturing model, for better or worse, has quickly taken hold in solar tech. Evergreen Solar had a technical advantage but, in the face of rapidly declining prices for solar-grade silicon, its proprietary technology was ultimately its undoing. Even moving manufacturing to China and adjusting its product didn't allow it to keep pace with price declines.
"Back in 2008, the future looked pretty bright. Evergreen had a patent on making (silicon) wafers more cheaply than anyone else," said Brett Prior, a solar analyst at GTM Research. "But by the fourth quarter of 2010, the Devens plant was facing a completely different industry. Modules prices had fallen 55 percent in two and half years--very few industries can cope with that."
Evergreen Solar is certainly not the only company to suffer financially from the onslaught of falling panel prices, which have already fallen 20 percent in the first half of this year. Germany-based Solon this week said it is closing an Arizona panel factory and will trim its workforce in Germany, and BP Solar shuttered its panel factory in Maryland last year. Germany's Q-Cells earlier this month warned of a large quarterly loss amid a restructuring to lower costs.
All companies paying more to operate in countries with high costs will likely face the same fate, say analysts. In that way, Evergreen's slide was a product of both industry trends and its own particular problems.
Evergreen, though, suffered so dramatically because it had both a high cost structure and a technology that made the company a small island in a sea of standard component manufacturers. To trot out the PC metaphor again, it would be like asking manufacturers to use a technically elegant but proprietary chip when the industry had standardized on the Intel's x86 processors. In fact, Sun Microsystems did exactly that for years and was constantly criticized for it.
The technology behind Evergreen originated at the Massachusetts Institute of Technology, where Professor Ely Sachs developed a method for making polycrystalline silicon wafers with about half of the material needed in traditional processes. A few years ago, that was a significant advantage as there was a global shortage of silicon-grade silicon, leading to price increases. Those square silicon wafers are turned into solar cells and then assembled into a panel.
The problem was that the wafers Evergreen turned out were not a standard size, which would mean a solar cell or panel manufacturer would need to have a separate production line just for Evergreen wafers. Even if it lowered the cost, that sort of technology lock-in scared off manufacturers.
"If any of the big players bought wafers from Evergreen, they'd be held over a barrel. It was a huge uphill battle," said GTM Research's Prior.
As a result, Evergreen made its own panels and sought to adjust its process to make standard 5-inch-square wafers. At 2008 prices of about $4 per watt, the strategy worked. But now with prices nearing a dollar and a half per watt for the lowest-cost producers, the math just didn't work.
That holds true for any other solar manufacturer with conventional technology trying to compete on cost alone. And per watt prices are a moving target.
"It's a tough landscape now for suppliers. As long as you remain uncompetitive on price and don't have some differentiation or serve a niche market, it's going to be tough to figure out how you compete," said Shayle Kann, a solar industry analyst at GTM Research.
Thin film advantage?
There are many more U.S.-based venture-backed solar companies vying to break into the big leagues. Among them include Miasole, Suniva, Alta Devices, Abound Solar, and 1366 Technologies, which was co-founded by MIT's Sachs. General Electric's relaunched solar business is based on technology developed by thin-film solar start-up PrimeStar.
Thin-film solar cells, which use a variety of materials, are less expensive to make but the efficiency is lower than traditional silicon cells. That means thin-film panels are typically well suited for utility-scale plants or commercial rooftops where space isn't a premium. Thin-film processes are also more automated so labor doesn't figure in as much in the final cost, noted Zindler.
But impressive technology alone doesn't translate to success, as Evergreen found out. One of the biggest advantages that Chinese manufacturers have is access to capital, say analysts. They also benefit from cheap access to land, low interest rates, lower fees for environmental compliance, lower cost of energy, and lower cost of materials, noted Prior.
For many solar companies, getting financing is crucial to scaling up and getting a chance to compete on the global stage. Silicon Valley-based SunPower, for example, sold a 60 percent stake to French energy giant Total to get access to $1 billion on Total's balance sheet.
For less established companies, the Department of Energy's loan guarantees, state tax breaks, and institutional investors appear to be the best route for financing. The DOE has conditional commitments for three loan guarantees around solar manufacturing to Abound Solar, Calisolar, and 1366 Technologies.
For policymakers and taxpayers, government aid comes with inherent risks, particularly in the manufacturing end of the cut-throat solar industry. In the case of Evergreen, Massachusetts state officials defended their decision to provide tax incentives. It also has a "clawback" clause where the state can seek to recoup some of the money Evergreen already received.
"It is a cautionary lesson. We knew that it would be challenging to do that kind of manufacturing in the United States," Massachusetts Economic Development Secretary Greg Bialecki told the Boston Globe. "We will take a hard look at whether the way we designed the agreements or the way we created the clawbacks [to recover money] worked."
Whether the latest crop of U.S. solar companies can compete by manufacturing in the U.S. is still an open question. Abound Solar and 1366 Technologies, for example, plan to build demonstration facilities in the U.S. and then larger-scale plants meant to bring costs lower.
Not all the economic value in the solar industry is tied to manufacturing, though. The supply chain includes upstream companies, such as silicon suppliers, as well as downstream companies that do installation, permitting, and operate solar farms, noted Kann.
In the meantime, political debates will continue over funding for solar research, commercializing nascent technologies, and subsidies for power production from renewable sources, which cost more than fossil fuel generation.
GTM Research's Prior argued that Evergreen Solar gave the state what it wanted, which was jobs. At the time of the state agreement, it looked financially solid as it was making more than $100 million in revenue and had a market capitalization over $2 billion.
But solar companies and policymakers need to consider domestic solar manufacturing with open eyes and understanding of the fast-changing industry dynamics. "When you're trying to get manufacturing companies to succeed and you give them incentives, it's a chance that it's not a local game. They're competing in a global market," Prior said.