Your daily dose of photovoltaic technology developments and solar news
Capacity conundrum: Are thin-film PV companies at a disadvantage versus the cheap silicon crowd?
Most Popular Articles
- Q-Cells CEO exits amid investor anxiety
- Moser Baer achieves 7.3% module efficiency through its proprietary process
- Yingli Green to add 300MW of capacity; tops $1B in annual revenues, achieves record module shipments
- Solar photovoltaic market grew 6% in 2009, says Solarbuzz
- Project Focus: SunPower scores 200MW rooftop PV contract with Southern California Edison
Popular Technical Papers
Subscribe to our weekly newsletter
The eyes and ears of PV market watchers will be on First Solar Wednesday afternoon, Dec. 16, as the thin-film kingpins deliver their 2010 guidance Webcast. Aside from the stock market practitioners’ hand-waving over how the update will effect their models and forecasts, one element of keen interest to th
e photovoltaic manufacturing community is whether the company that made CadTel famous will announce plans for adding production capacity to its existing 1GW base, specifically in North America, where the utility market’s siren song certainly has First’s attention across its internal value chain.
In his comprehensive preview note issued Monday, Barclays Capital Clean Tech analyst Vish Shah weighed in on this, discussing the possible impact on First’s cost roadmap if it decides to set up more factories in the U.S. As he points out, labor would be more expensive than in Malaysia, but cheaper shipping costs to U.S. projects and higher conversion efficiencies would ameliorate that labor cost. There’s also the question of “forward demand visibility,” and how the muddled view may make the solar company hesitant to pull the trigger on building out capacity (except via organic growth through improved throughput and efficiencies on existing lines).
But a thought-provoking nugget pops out about halfway through Vish’s analysis that has more to do with a broader issue facing thin-film companies as they map their manufacturing strategies and stare down the challenge of cheap silicon modules from China. His comments reside under the subhead: “Do Thin Film Companies Have Capacity Disadvantage Over Silicon Companies?”
Here's what Vish (pictured below) wrote:
“Underutilization impact is greater in the case of thin film companies than in the case of silicon companies that either focus only on a part of the value chain or have a flexible vertical integration/manufacturing strategy. Given the demand uncertainty in the subsidy markets as well as greater variability of peak demand, solar companies have to plan for some flexibility in capacity (capacity plans of Trina, Canadian Solar are good examples of the new environment). 
“This flexibility may not be available for thin film companies. Lower utilization rates lead to higher costs (example Energy Conversion Devices) and conservative capacity planning leads to missed revenue opportunity (First Solar could have shipped more in Spain in 2008, Germany in 1H10 if it had more capacity). In both cases, thin film companies seem to have a capacity disadvantage over silicon companies, a factor that could influence market share dynamics of thin film companies over time.
“First Solar's current manufacturing strategy, which is based on maximizing utilization rates, is perhaps the right strategy for all thin film manufacturers. Moreover, we agree that the US utility market is the best option to secure the demand visibility and plan additional capacity expansions over time. However, until the US market pipeline grows to become more meaningful compared to European market demand (we forecast this scenario in 2012-13 timeframe), risk of market share loss to the more aggressive Chinese companies exists in a low poly(silicon) price environment.”
I wouldn’t be too concerned about whether First Solar will be adversely affected by this dilemma, given the executive team’s laser focus, hard-nosed business savvy, and long-term outlook. But I do wonder about the implications for the scores of thin-film PV companies—be they amorphous silicon, CIGS, or CdTe--either ramping their first production facilities or adding a second line or facility.
Few are running even close to full capacity and many are struggling to get those sub-buck-a-watt cost efficiencies that only scale and volume can bring. If they do stumble, prognostications of thin film taking over increasing chunks of market share from the crystalline crowd could require an adjustment—especially for TFPV without the First Solar brand.
Cart
There appears to be nothing in your cart!
Historically, scale has been the numero uno factor driving costs down. The research is solid. So uh... yeah, you can assume scale is the most critical factor overall. That's a safe bet. The wafer guys say that. The inverter guys say that. You see that in installation. We'll see that in silicon production more and more. Will you always win betting on scale? Obviously not. Do technologies matter? Absolutely. But... if you had to 'play the odds' bet on scale driving down costs more than anything else. Why fight history?
Obviously Vish has not done his research very well. FSLR ability to replicate production facilities is phenominal.
It is wrong to assume that scale and volume are critical for "sub-buck-a-watt cost." The critical element is technology (specifically, efficiency and yield). Here is the proof: In 2005, First Solar produced 21MWs at $1.60 per Watt cost of manufacturing. In the same year, Unisolar produced the same amount of MWs but at a manufacturing cost of $2.60 per Watt. That $1 per Watt disadvantage for the inefficient, inferior and outdated technology (circa 1977) is simply impossible to overcome, regardless of the scale or capacity utilization. And that is why First Solar produced 292MWs in the September 2009 qtr at 85c per Watt, while Unisolar could just do 33MWs at $1.76 per Watt and could not sell them (unsold inventory reached 42MWs, and that's before all the MWs of unsold stuff they reacquired by "buying" their largest customer). In the December 2009 qtr, Unisolar is guiding for horrible numbers (shipments of less than 19MWs and cost of manufacturing over $2.20 per Watt). Anybody who wants to compete with First Solar needs to have cost of goods manufactured that are equal or better than First Solar's on efficiency-adjusted basis - meaning, a 6%-efficient panel has to have costs of manufacturing of less than 60c per Watt, and a 9%-efficient panel needs to have cost of manufacturing of 75c per Watt or less, and a 15%-efficient (crystalline) panel needs to have cost of manufacturing of about $1.35 per Watt. Only players that have a technology with such a cost structure will be able to grow and benefit from the virtuous cycle of scale and utilization. Unfortunately, no thin film player, other than First Solar, has demonstrated such a technology (neither Applied Materials, nor Oerlikon, nor Ulvac, nor Sharp, nor QS Solar, nor the new crop of various CIGS players). The "capacity disadvantage" is just a secondary consideration. First Solar will produce more MWs worth of panels than any other company this year. Thus, apparently, they still compete effectively with crystalline.