Paula Mints from Navigant Consulting looks at long-term and near-term demand trends, to illustrate what the PV industry needs to develop a sustainable market vs. the perception of being a highly subsidized type of commodity: electricity.
From 2004 through 2009, the photovoltaic industry experienced a boom in growth, with demand increasing by a compound annual 50% for the five-year period, and industry growth at 44% in 2009 vs. 2008. Driving this extraordinary growth was Europe’s feed-in tariff (FiT) incentive model, and in particular the strong market in Spain in 2008, and the strong markets in Germany, Italy, and the Czech Republic in 2009. In 2010, strong demand in Germany, Italy, and the Czech Republic will drive annual growth of >65%.
The 2004-2009 period managed to encompass several significant events:
- Silicon feedstock shortage and significant high prices for this raw material
- Steady price increases from 2004 through 2008 for photovoltaic cells and modules
- ~50% price decrease in 2009 over 2008 for photovoltaic cells and modules
- Significant increase in market share for thin films, and in particular, CdTe
- The rise of the multi-megawatt (utility-scale) installation
- A global recession and the crisis in the financial sector that uncovered the a virtual shell game in the trading of derivatives that exposed significant housing market debt
- Constrained debt and equity markets
- The proliferation of the feed-in tariff incentive model globally
For a 35-year period (1974-2009), demand in the PV industry grew at a compound annual rate of 42%. Following the capping of Spain’s generous FiT at the end of 2008, demand slowed significantly. Consequently, demand during the first quarter of 2009 was flat, with high levels of inventory on the demand side, specifically in China. Complications to the market situation at the beginning of 2009 include poorly designed systems (primarily in Spain), poor performance of c-Si module product using polysilicon blend starting material, and the selling off demand side inventory at extremely low prices.
At the end of 2009, the German government, somewhat dismayed by close to 4GWp being installed in that country, announced strong additional decreases to its FiT, but failed to implement these changes until mid-July. Despite poor weather and because of the impending changes to FiTs in both Germany and the Czech Republic, the market in Germany surged in 2010. Other strong markets in 2010, including Japan and the US, have cemented accelerated growth in 2010.
Figure 1 highlights the metrics that made up 2009 and 2010 in the PV industry, including demand-side inventory, shipments, production, capacity, and installations. The figure separates 2009 and 2010. In 2010, a bar for estimated defective modules is included.
|Figure 1: 2009/2010 PV industry metrics.|
The photovoltaic industry remains immature, requiring incentives to stimulate demand. Its product is electricity, and the delivery of this electricity remains expensive. Moreover, the product can be defined as a commodity and as an appliance, and also as a necessity and a luxury item. A solar panel is the Cadillac or Mercedes or BMW of electricity.
With reliable and less-expensive substitutes available, a natural pull (demand without incentives to stimulate it) from end-users does not yet exist. Studying specific periods in the PV industry’s history in terms of growth and drivers for growth as what can be learned from these period can help us understanding the direction of this still young industry. Figure 2 offers compound annual growth rates for the PV industry for specific periods: 1974-1984, 1984-1994, 1994-2004 and 2004-2010.
|Figure 2: PV industry history, 1974-2010.|
What a difference a decade makes
Solar products are still sold into applications. Though the basic application categories have not changed application market share has changed dramatically in just a few years. In 1999, demand for (and sales into) the grid-connected application was 39% of industry sales — and the industry was <200-MWp. In 2009, the grid-connected share of total demand was 95%, and the industry was ~8-GWp. Table 1 lists historic market share by major application for 1999, 2004, 1998, and 2009 also presenting compound annual growth rates under the conservative and accelerated (ACC) scenarios for 2014 and 2019.
Table 1: Market share by major application historic and forecasted percentile values.
(Columns or rows may not add due to rounding.)
Table 2 provides an aggregate multi-megawatt large field or roof installation forecast from 2009 to 2014 for the commercial forecast. As CSP and CPV mature, the multi-megawatt commercial application will become significantly more competitive and PV will require higher levels of electricity production, which will drive the use of trackers. This demand will also drive improvements in tracker efficiency and reliability. In general, requirements from investors for larger more profitable systems, and the future commoditization of the sector will push large commercial towards multi-megawatt ground mounted configurations. Barriers to multi-megawatt installations include permitting, transmission, concerns about overuse of land in Europe and the US (particularly agriculture), the Bureau of Land Management in the US, financing, incentive availability, the opening of the markets in China and India (crucial to the accelerated forecast), component (module and inverter) availability, and system installation, EPC and investor expertise.
Table 2: Multimegawatt commercial estimate, 2009-2014. Large-field systems are defined as >1MWp.
(Columns or rows may not add due to rounding.)
Global markets for PV
Europe (at 80% of total sales) continued to dominate demand in 2009, just as China and Taiwan (at 46% of sales) dominated supply. Even with the mid-year and late-year changes to Germany’s FiT, Europe is set to continue as the primary global market for PV products in 2010 at 77%-80% of total. Strong demand from specific markets in Europe led to shortages of modules in other regions, and to shortages of inverters globally in 2010. From 2004-2009, demand in Europe increased by a compound annual rate of 68%. Demand in North America (Canada and the US) had a compound annual growth rate of 39% for the same period. The Middle East, specifically because of Masdar City, enjoyed extremely strong growth (>300%), which is not expected to continue. Table 3 presents regional demand growth from 2004 through 2009.
|Table 3: Regional demand growth, 2004-2009. (Columns or rows may not add due to rounding.)|
Table 4 provides regional demand statistics by application segment for 2009.
|Table 4: Regional application breakdown, 2009. (Columns or rows may not add due to rounding.)|
Concerning modules, size matters
The subject of module size is important in terms of product development for the manufacturer, though, the industry remains focused on one attribute of its product development — wattage. Except for Uni-Solar’s flexible product for rooftop installations (particularly metal roofs), flexible products have failed to gain significant share in the market. There is continued development of modules >250W, particularly for the commercial and utility markets, but also for the grid-residential application, even migrating to some remote applications. In 2007, 42% of modules sold into the PV market were >200Wp. In 2008, 48% of modules sold were >200Wp. In 2009 there was a significant increase in sales of modules >200Wp-300Wp, to 65% of total module sales. This change in buying behavior is significant as leads directly to a change in product development plans by manufacturers, leading to less availability of lower wattage modules. Figure 3 presents market share by module size for 2009.
|Figure 3: Market shares by module size, 2009 demand. (Power modules are all modules other than those used for the consumer indoor application)|
In 2009, sales to Europe represented 80% of global industry sales. In 2010, data suggest a one-percentage point increase in Europe’s market share to 81%. The significance of the incentive driven market in Europe on the global PV industry along with the vulnerability of this global dependence cannot be understated.
Weighted average module prices for select countries in the EU, along with the EU average are as follows:
- France: 2.02-Euro/Wp
- Germany: 1.90-Euro/Wp
- Greece: 1.80-Euro/Wp
- Italy: 1.84-Euro/Wp
- The Netherlands: 1.93-Euro/Wp
- Spain: 2.00-Euro/Wp
- Switzerland: 2.37-Euro/Wp
- Czech Republic: 1.75-Euro/Wp
- Portugal: 2.13-Euro/Wp
- Europe ASP: 1.97-Euro/Wp
Average system prices in Europe in 2009 were as follows:
— Residential: Average system size 5.5kWp, 5.71 €/kWp, expected to decrease by 7% in 2010
— Commercial: 4.78 €/kWp, expected to decrease by 12% in 2010
— Utility Scale (multi-megawatt): 3.10 €/kWp, expected to decrease by 4% in 2010
And finally, into 2011 and beyond
Continuing concerns over the environment, volatile energy costs, and a perception of declining reliability under a restructured utility infrastructure will have a significant effect on demand for grid-connected PV products in future years. The strongest drivers for the grid-connected commercial application are incentives — particularly the feed-in tariff model, which enables system owners to sell solar generated electricity back to the utility or to end users. Investment business models that manipulate available incentives and system component prices for efficient system installation and profitable solar-generated electricity sales began with Europe’s FiT incentive model.
Although this is not intuitive, near- to middle-term demand for PV systems will remain volatile as incentive schemes prove unsustainable, and substitutes emerge such as CSP for utility-scale applications. Proponents of nuclear technology continue to make its case as a renewable technology, and cheap, polluting coal continues to take a significant share of energy production.
Concerns for the future direction of PV include branding a product that is, basically, a commodity (electricity). For the residential application, when the end user buys the technology — and not just the electricity — solar will have made a significant step forward to developing a sustainable market. The analogy in this regard is the automobile. In most cases the choice of a car is more complex than simply buying a mode of transportation. As the PV industry matures it will need to market itself on various buying behavior levels to all stakeholders and end users, while working to help brand names emerge in the consumer consciousness.
Despite significant or otherwise changes to the FiTs in Germany, Ontario, the Czech Republic, and France (among others); despite Fannie and Freddie placing a Pace-denying roadblock in front of residential solar; despite competition (particularly in the developing world) between the FiT and tender process incentive models; despite the risk that the ITC grant may not be extended; despite … despite … despite — 2011 will be a solid year for the PV industry. The estimate for 2010 is, of course, the accelerated scenario, with accelerated growth likely in 2011 also. After all, there is significant manufacturing capacity — and what’s a constrained margin or two among friends?
|Figure 4: Lowered incentives, conservative, accelerated forecast to 2014.|