The solar industry is a very interesting industry, even if it's somewhat frustrating for investors. A few characteristics:
The production process isn't very complex.
The production process displays moderate economies of scale and learning (reducing cost with scale and experience).
The industry displays low barriers to entry.
The product is a near commodity.
There are nevertheless several different technological approaches, with many more approaches presently sitting in labs or in the experimental stage.
The market grows at a very rapid pace, and can reasonably be expected to continue to do so.
The market nevertheless depends to a large extent on the existence of subsidies.
However, it can be reasonably expected that subsidies will no longer be necessary in a growing part of the world.
Solar energy has some inherent disadvantages stemming from the unpredictability of the source.
This is basically how the market looks today. The enthusiasts point to the rapidly growing market and the reduced need for subsidies due to large falls in 'average selling prices' (ASP). They also point to the bright future when the industry can do without subsidies altogether, having reached the mythical 'grid parity' in most places (that is, solar energy being as expensive as electricity from the grid).
All of that is true (even if 'grid parity' is a rather hazy concept). However, there are also negatives, especially for investors. The fact that the product is a near commodity, together with the low entry barriers, make most companies close to the proverbial 'price taker' in the perfectly competitive markets one marvels at in economics 101.
The lack of entry barriers, existence of take or pay supply contracts and a moderate amount of fixed cost explain why the industry is plagued by bouts of over-investment and oversupply, leading to dramatic price falls and inventory write-offs, or even industry shake-outs. We're at the tail end, hopefully, of just such a period. While these price falls stimulate market growth and take us ever nearer to grid party for ever more markets, they take years in market expansion for corporate finances to recover from such onslaught, as last year's near 50% ASP reduction showed.
If there is a way out, it is likely through diversifying and/or distinguishing the company's product in one way or another. It could, like Suntech (NYSE:STP), use premium materials in order for its product to command premium prices. Results from Suntech suggest it doesn't yet make much of a difference, but perhaps it will.
Others, like Trina Solar (NYSE:TSL), try to ramp up new products and services (its honey cells and Trina mount, for instance). Like Yingli (NYSE:YGE), Trina has also integrated backwards, but has a buffer of suppliers high up the supply chain for flexibility. As a result, Trina is generally considered to have the most efficient production process, leading to somewhat higher margins compared to the competition.
LDK (NYSE:LDK) is completely integrated in the sense that it also operates polysilicon plants. Trina was supposed to go this way as well, but ultimately decided against it. Others, like First Solar (NASDAQ:FSLR) use whole different technologies that afford even more automated production processes. This was very good while polysilicon prices (the ingredient of the main competition) were high, but now that these have crashed to earth, it isn't so good. And the automated factory offers limited scope for further cost cutting.
Jinko Solar (NYSE:JKS) has basically used another textbook strategy, more or less copying Trina Solar's approach and even hiring a successful Trina sales director.
Apart from process efficiencies and branding, all the main solar companies focus on reducing the amount of materials used ('thin films' becoming ever thinner) and increasing conversion efficiencies, the amount of solar light that the cells manage to convert into electricity. Here is an overview of recent results by some companies:
Suntech's Pluto cell technology achieved 20.3% efficiency, with 21% targeted next year.
First Solar , the U.S. thin film king set a new world record for the CdTe technology with 14.5% efficiency.
Alta Devices (not public but backed by famed venture capitalist firm Kleiner Perkins Caufield & Byers, amongst others) uses GaAs (gallium arsenide) technology which reached a 23.5% efficiency.
SunPower (SPWRB) has a history of setting efficiency records, its new Maxeon cells exceed 22% conversion efficiency.
MiaSolé, the third producer in the thin film sector using the CIGS technology has announced a 17% efficiency cell.
Solar Junction (private) broke the world record for concentrated solar power (43.5%) by concentrating solar rays onto cells.
- Abound Solar (private) produces CdTe (cadmium telluride) modules which have achieved a 12.5% conversion efficiency, just under those of First Solar which uses the same technology.
These figures are not really comparable though. Different technologies have very different cost pictures, and some are lab results, while scaling up to commercial production levels often leads to some efficiency losses.
For existing players, the present market conditions are not the only thing that matter. While they continue to shave off valuable materials, improve production processes and make cells more efficient, material scientists, labs and start-ups around the world are embarking on commercializing whole different approaches and materials.
If one of these becomes successful, this could undermine the business model of existing players. We already provided something of a first inventory, here are some more promising new approaches.
Natcore Technology controls a new thin-film growth process that promises to allow mass manufacturing of tandem solar cells with twice the efficiency of the best solar cells available today. This would mean that solar energy would finally be cost-competitive with conventional power. And that we can significantly diminish our dependence on fossil fuels.
The company's joint venture in China has even become a case study at the Harvard Business School. The patented process, developed at Rice University, involves depositing "coatings in a low-temperature, liquid-based process rather than the high-temperature gas-based process" (MIT Technology Review).
But this is not the only route they're taking as the company is also involved in developing solar cells from carbon nanotubes or nanoscale crystals ('quantum dots'), technologies for which commercial applicability is still years away.
The company plans to earn though licensing, rather than manufacturing its own cells. So far, its $6M IPO isn't much of a success though.
Adam Stratton, Power Panel's vice president of Business Development, explained that the panels maximize the sun's energy by capturing 80% rather than the typical 5-18% captured by a traditional photovoltaic (PV) panel. This is important because, according to Stratton, a typical home uses roughly 70% thermal energy and around 30% electricity.
MIT Labs have created a stacked solar panel system that "produce up to 20 times the power output of conventional solar power installations" [extreme tech]. The approach is based on the fact that solar cells themselves only constitute some 35% of overall installation cost. The main issue is the low energy density and the sheer surface area necessary to generate the electricity. By stacking cells, a much higher density can be achieved.
MEI's Pura is a processing technology (called Pura) to clean polysilicon, allowing higher levels of conversion efficiency and they're signing up customers in quick succession (even from the semiconductor sector). Here is Ed Jean, MEI's global sales manager:
"What's driving solar right now is who can have the highest efficiency for the least cost," Jean said. "It has become critical that the raw materials have higher and higher purity."
Twin Creeks Technology, a formerly secretive company finally had a coming out party at which it showed off equipment that enables solar module makers to significantly reduce the amount of silicon used. This matters a great deal, as silicon accounts for 20-25% of the cost of a PV module. At present, wafers have silicon layers 150 micron thick, but with Twin Creeks technology, this can be greatly reduced to 20 to 30 microns.
But they are not the only one with these kind of radical improvements. Ampulse, AstroWatt, Bandgap Engineering, Crystal Solar, Silicon Genesis 1366 Technologies all have or are developing processes which enable much thinner silicon layers.
Halotechnics, a solar-thermal startup is innovating around two of solar energy's most intractable problems. The sun doesn't shine at night (or on cloudy days) which makes energy storage a rather big target.
The company has developed new heat-storage materials that promise to not only make solar-thermal power plants more efficient, but also reduce the cost of storing energy from the sun for use when it's most needed. [MIT Technology Review]
Solar thermal plants work differently from solar panels. The latter convert sunlight directly into electricity while solar thermal plants concentrate solar light with the help of mirrors on water basins that power turbines, generating electricity indirectly. The advantage this approach has is that the heat is easier to store (during the night) than the direct electricity generated by solar panels), even though the electricity itself is more expensive.
This is only a small sample of a world of innovative ideas and companies that could easily shake up the industry. Existing solar players do not only have the difficult market circumstances to worry about.
Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.