The solar photovoltaic (PV) industry suffered a major blow last year when the Spanish market collapsed, but despite this loss the market performed much better than expected. Shipments to the first point of sale, including installers, system integrators, distributors and other technology manufacturers, grew by 44% to 7.9 GW. This was a significant rise from 5.5 GW in 2008.
The year began with practically non-existent sales from manufacturers as the industry worked off more than 1.2 GW of demand-side inventory. By the middle of the second quarter, however, the inventory had been completed and demand in Germany, France, Italy and the Czech Republic took off.
|An industry that relies on incentives... and suffers criticism for this reliance, is always risky to enter or invest in.
Shipments do not equate exactly to installations but last year's installation figures are impressive: Germany totalled around 3.8 GW, Italy around 1 GW and the Czech Republic close to 500 MW. When Japan's new feed-In tariff (FiT) began in November, demand in that country grew close to 500 MWp. As a result, the 2009 market for PV technologies came as an overall relief to the global PV industry.
Looking back at 2009, there are many different numbers to analyse and just as many categories to put them in. Figure 1 shows these categories, including inventory, shipments, production, capacity and installation for the year.
PV industry growth over the past five years has been impressive by any standard, with two caveats; the industry remains entirely incentive driven for grid connected application and grid connected application represents 95% of industry demand. Until the industry can function without incentives and subsidies it will always be a risky market to invest in or to enter as a manufacturer or installer. (Of course, all energy technologies are subsidised, including conventional energy).
Figure 2 shows PV industry growth from 2004 through 2009. During this period the industry grew by a compound annual rate of 50%.
Shipments grew by an impressive 44% compared to 2008, which was particularly surprising in a year when they had been forecasted to decrease. Unsurprisingly, given the low prices from manufacturers in the region, 47% of 2009 shipments came from China and Taiwan. Table 1 presents shipment information from 2004 to 2009.
During this period, shipments from China and Taiwan grew at a compound annual rate of 145%. The ROW region includes manufacturers from India, Malaysia and the Philippines.
Last year, strong demand coincided with aggressive pricing from manufacturers in China and Taiwan. In mid-2009 it was cheaper in many cases to outsource cells and modules from China or Taiwan than it was to manufacture them in-house. Outsourcing and tolling (for manufacturers with wafer capacity) became commonplace, while prices dropped significantly. It was a good time for companies buying technology but not so good for technology manufacturers. Thin-film faced a particularly difficult competitive position with low-priced crystalline driving margins to uncomfortable levels.
Average selling prices – 2009 from 2008
- Thin-film fell 45% from US$3.00/Wp to US$1.65/Wp;
- Cell fell 60% from US$3.20/Wp to US$1.26/Wp;
- Module for large quantity buyers (typically 50 MWp to over 100 MWp a year) fell 33% from US$3.25/Wp to US$2.18/Wp;
- Module for mid-level buyers (typically 10 MWp to over 25 MWp per year) fell 23% from US$3.65/Wp to US$2.82/Wp; and
- Module for small quantity buyers fell 27% from US$5.02/Wp to US$3.68/Wp.
Table 2 presents average selling prices (ASPs) over time for thin-film technologies and large quantity buyers of all technologies. The large quantity buyer category is representative of crystalline technologies. PV technologies with lower conversion efficiency have an area penalty (this includes higher balance of system and installation costs) that requires a lower price/Wp. Typically this price must be lower than the lowest available price for crystalline technology.
CSP versus CPV
2009 began as a year of dire predictions and low expectations, and ended with strong demand, bargain basement prices and stressed margins for the PV industry. So what did it mean for concentrated solar power (CSP) and concentrated photovoltaics (CPV)?
Size is crucial in the case of CSP. Installation needs to be in the megawatt range to make economic sense, while flat plate PV can be installed on roofs and in smaller kilowatt configurations. The solar world has, however, been leaning towards multi-megawatt (or utility scale) installations for some time. In this regard, CSP projects have the clear advantage of storage. Utility companies, in particularly, continue to be concerned about the variability of solar energy and are very interested in the storage capabilities of CSP.
|Shipments to the first point of sale, including installers, system integrators, distributors and other technology manufacturers, grew by 44% to 7.9 GW. This was a significant rise from 5.5 GW in 2008.
The primary issue facing the CSP community these days is that crystalline silicon (c-Si) flat plate is currently cheap and has a more consistent track record in the field. A utility-scale PV installation with tracking will be up and running faster and has lower operation and maintenance costs than CSP. Both technologies also face problems of insufficient transmission availability, trials and tribulations with the Bureau of Land Management in the USA, slow recovering global debt markets and, in the case of large fields, the need to deploy on large swathes of flat land.
CPV continues to face a rockier road. During the 2004 to 2008 boom (which coincided with a polysilicon shortage) CPV had one significant advantage – the technology uses less or no silicon. During the polysilicon constrained years, when prices for raw material rose to over US$400/kg on the spot market, investment in CPV start-ups was strong. CPV technology development, however, is a delicate balance.
The sector has the theoretical advantage of low costs and high efficiency but flat plate PV continues to improve in these areas. CPV technologies require highly efficient tracking and a perfect balance of optics and technology; it is a high wire act with significant potential but requires perfect timing.
Forecasts for the coming years
2010 has started out with welcome price increases for manufacturers of wafers and cells, along with extremely strong demand for PV technology in Germany, Italy, the Czech Republic and France. The industry is currently shifting everything it can produce to those countries, particularly Germany. This is having dire consequences for the market in Spain and will eventually have a significant impact on the German market too.
However, increasing prices are easing margins and this is good news for technology manufacturers. Figure 3 presents average technology prices from 1989 through an estimate for 2010 for large quantity buyers, along with a weighted average for all buyers (reflecting different prices for markets and buyers.) Figure 4 presents technology revenues from 2004 to 2009 (when the industry sold more and made less), along with revenue forecast through to 2013.
Overall, 2010 should show significant growth for PV up to at least 11 GW, with Europe again the leading market. Thin-film technology, in particular, may have an opportunity to recover despite investors believing it is a riskier investment. In general, the venture and investment community may be starting to realise that all photovoltaic technologies require a longer gestation period until they are commercially viable compared to other investment opportunities, and that even once they are commercialised it takes time to achieve healthy annual sales. The PV industry is indeed maturing but it will not take off overnight.
|Size is crucial in the case of CSP. Installation needs to be in the megawatt range to make economic sense, while flat plate PV can be installed on roofs and in smaller kilowatt configurations.
Reliance on incentives
With strong growth for PV certain in 2010, where does it go from here?
As long as the industry requires incentives its future is hazy at best. Figure 5 provides a PV forecast to 2014 for three scenarios: reduced incentives, conservative and accelerated. The primary difference between conservative and accelerated forecasts is the availability of incentivised markets. The accelerated forecast assumes continued growth in markets such as Germany that, although they appear saturated, continue to consume products.
Every indication coming from the German Government seemed to show that significant changes to its FiT would have happened by now, including an end to incentives for agricultural land. Although it would seem a safe bet to assume that these changes will eventually come about, there are no sure-fire certainties in solar. One thing is for certain; if and when Germany and other countries alter their FiTs, average technology prices will decrease and margins will once again be compressed.
|Despite the hazy oultook for CPV and CSP, flat plate PV will have a good year with shipments of around 11 GW, slightly higher prices meaning more comfortable margins and the high possibility of a healthier investment climate.
CSP installations (trough, power tower and dish) have a much longer timeline than flat plate PV. In general, it is necessary to assume that it will take two years from the start of construction. This long timeline is problematic in terms of forecasting and CSP is unlikely to ever provide a smooth trend line in this regard.
Table 3 offers an overview of announced and installed CSP systems to 2015. Note that one fifth of the total 6.2 GW have an estimated completion date of 2010, but it is highly unlikely that it will be finished this year. Instead, 4% of the 2010 installations will likely be completed and the remainder pushed forward or cancelled. The 24% of systems with no announced completion date are also problematic.
Despite the hazy outlook for CPV and CSP, flat plate PV will have a good year with shipments of around 11 GW, slightly higher prices meaning more comfortable margins and the high possibility of a healthier investment climate.
This does not mean that PV's problems are over; an industry that relies on incentives to such a high degree and that suffers significant criticism for this reliance, is always risky to enter or invest in. However, given over 35 years of positive growth and the accelerated growth of the past five years, it continues to surprise and reward its participants.
About the author:
Paula Mints is the principal analyst for Navigant's PV Service Market Research Program, executive editor of the Solar Outlook Newsletter, and Director of the Energy Division.
The PV Services Department at Navigant Consulting was founded in 1974 at Strategies Unlimited, and Ms Mints moved it to Navigant in 2005. The practice is based on classic market research principles; that is, all data are primary, not secondary, and the analysis is independent and not based on the work of others.