Low margins for photovoltaic manufacturers have a history: until 2004 most PV manufacturers had low to no margins on cell and module sales. The price function has historically been controlled by demand participants (system integrators, installers et al) and this group was not price elastic – nor, as their own demand function was a struggle, did they need to be.
Pricing in the solar industry has traditionally only had a glancing relationship to cost. Before 2004 aggressive pricing situations were common (for example, 2000 through 2003), with the sole purpose of gaining share towards the day that demand would boom.
Germany’s Renewable Energy Act, the EEG, changed the low-to-no-margin paradigm within which PV manufacturers did business, though change did not happen overnight. Germany’s solar incentives progressed steadily from capacity-based to the modern feed-in tariff (FiT), defi nitely not at lightning speed (except through the prism of memory) and with the end result hard-fought by many, including German MP Hans-Josef Fell, father of the modern FiT. 2004 began a period of increasing margins for PV manufacturers despite a polysilicon shortage that led to extremely high spot market prices for c-Si necessary start up material.
Figure 1 (above) offers pricing data from 1992 through 2012. These data are based on annual survey effort that dates to the mid-1970s and represent average global module selling prices to the first buyer. The fi rst buyer in the market can be another manufacturer. Figure 1 includes the 2012 average selling price (ASP) for inventory. Prices for inventory confuse the overall pricing data. Despite the signifi cant decrease in module ASPs (to the fi rst buyer) decreased by a compound average of 9% for the 20-year period refl ected in Figure 1.
Figure 1: Average Module Selling Prices, 1992 – 2012
During the most recent fi ve-year period module ASPs decreased
In an historic view of industry pricing and shipment/demand
growth from 1975 through 2011, we may compare periods of high and low incentives with market development during these periods:
1975-1985: The environment during his decade can be considered high incentive. During this period ASPs decreased by a compound average of 21% while shipments/demand grew by a compound annual 69%. 1985-1995: The environment during this decade can be considered low incentive.
During this period ASPs remained relatively fl at, decreasing by a compound average of 3%, while shipments/demand grew by a compound average of 14%. 1995-2005: This decade began with a low incentive environment transitioning to a high incentive environment by the end of the 1990s. During this transition decade, prices decreased by a compound average of 3% with shipments/demand growing by a compound average of 35%. 2005-2011: During this transition period the industry began moving from a high incentive environment to a low incentive environment.
ASPs decreased by a compound average of 15% with shipments/demand experiencing a CAGR of 60%.
Economic market theory holds that strong demand typically leads to periods of price increases. Realities are always more complex than are theories. Industries such as solar that operate in an environment with many substitutes and, in particular, substitutes that enjoy accepted subsidies and with the expectation of low prices, are not in control of the pricing function. Aggressive pricing for shares in such a vulnerable industry environment can, and did, lead to bubble-like behaviour.
THE HERE, THE NOW, THE LOW MARGINS Figure 2 (see page 22) highlights 2012 quarterly ASPs to the fi rst point of sale, for high effi ciency technology to the fi rst point of sale and inventory ASPs.
Unlike the historic period, when PV manufacturing was primarily supported by large multinational organisations or utilities, current manufacturers have primarily (with some exceptions) chosen to be self-suffi cient. In just a few recent years the rush to public company status opened the curtains on the mechanics of an industry that was probably too young to support such visibility.
Despite the weakening viability of many public solar companies, downward price pressure towards the end of achieving grid parity has been particularly harmful. As conventional energy technologies can and will likely continue to enjoy subsidies, the achievement of grid parity is unlikely to prove a bellwether of success.
Current low prices for PV technology (cells and modules) have driven photovoltaic technology manufacturers either to the brink of insolvency or over the brink. In some cases, such as Hanwha’s acquisition of Q-Cells, innovation will continue. In other cases, Uni-Solar and Evergreen for example, the potential for innovation in manufacturing costs, conversion effi ciency and product development had come to an abrupt stop. Industry pioneers, Schott Solar for example, have exited.
Remaining manufacturers in most cases have pragmatically chosen to cancel expansion plans and idle existing lines, though a few have chosen to ramp up despite mounting evidence that prices will not increase or simply stop falling in the near to mid-term.
Figure 3 (see page 23) presents ASPs to the fi rst buyer, inventory ASPS along with First Solar’s published manufacturing costs. Using First Solar as an example, the data in Figure 3 serve to highlight the diffi cult competitive position faced by thin fi lm technology manufacturers in the current pricing environment.
During the period depicted in Figure 3, ASPs to the fi rst buyer decreased by a compound annual 25% and inventory ASPs decreased by a compound annual 30%. CdTe technology manufacturer First Solar, for several years the acknowledged low manufacturing cost leader, decreased costs by a compound annual 7%.
The learning curve progress experienced by First Solar during this period follows a normal pattern and is indicative of progress, whereas the steep decreases of fi rst buyer ASPs is indicative of a market out of pricing equilibrium.
2013: ANOTHER ROLLERCOASTER YEAR Shipment activity ticked up at the end of 2012 (particularly in December) for at least two time-honoured reasons:
a) As the end of the year approached manufacturers wanted to rid themselves of inventory at any price (inventory is a cost);
b) The downward slide in prices is irresistible and is assumed to be a better hedge against future price increases than waiting for reductions.
A short-term increase in shipment activity may or may not be indicative of industry recovery or increased growth, and is simply a behaviour that should be observed over time. A true recovery for the solar industry would include margin recovery and there is no indication that this is imminent.
At the beginning of 2013, capacity to manufacture technology is at ~35 GWp, with module assembly capacity about 1 GW higher.
Figure 3 (see page 23) offers a snapshot of PV industry metrics in 2012 moving into 2013. The state of the industry in 2012 set the landscape within which 2013 will evolve.
For example, at the end of 2012 there was signifi cant inventory on the demand/supply sides of the PV value chain. This inventory will hold prices down in 2013. Installations in
2012 were higher than shipments because of the preceding year’s inventory lag. Historically in the PV industry, most years have ended with inventory.
In the aggregate, supply and demand data operate much like an accounting ledger in that both sides of the ledger (debit and credit) must equal at the end of the period. Essentially, if a megawatt was shipped, it was purchased. That same megawatt may not be installed in the calendar year it was shipped, and it may be resold. Though in the aggregate shipments and demand equal, each market exhibits its own behaviour.
Europe and the US have become import markets, while manufacturing in the two regions is slowing. Japan and China continue as export markets with strong domestic installation bases. Shipments and demand for Europe, the US, Japan and China from 2006 through (an estimated) 2012 were defi nitely not equal.
Figure 2: 2012 Quarterly ASPs
European demand rose steeply as shipments plateaued, while in China shipments climbed steadily, as did demand after spiking sharply in 2010.
Rumours of low prices are inappropriate vehicles on which to base a price strategy. Currently, the PV industry’s pricing behavior is volatile with prices held down and decreasing due to high levels of inventory, high levels of capacity, buyer expectations of low price levels and continuing decreases, and low incentive levels. There is an emerging trend to set PPA/FiT prices by tender for the markets for utility scale/multi-megawatt installation. This trend encourages buyers for this category to insist on the lowest price for hardware.
Given the market conditions, buyers have control over the price paid for modules with the caution that long term contracts, never stable instruments in PV, are highly unstable for both buyers and sellers under the current circumstances.
Given the current conditions of, again, high levels of inventory and capacity and a changing incentive landscape, price forecasts are highly unstable. In sum, the industry environment is not appropriate for spot price forecasting unless care is taken to defi ne which price point (inventory, fi rst buyer, high effi ciency, and thin fi lm) is being studied. Other categories include cell technology manufacturer to module assembler and onward, and distributor to secondary buyer, and buyer levels (small to multi-MW).
Figure 3: First Buyer Inventory ASP & First Solar
Figure 4: 2013 First Buyer ASP & Inventory ASP Estimate Each designation exhibits a specifi c buying behaviour. Currently, all buyers can access the lowest prices. Figure 4 offers an estimate of 2013 ASPs for fi rst buyer and inventory modules. The accelerated forecast currently has the highest probability of accuracy.
ONCE MORE UNTO THE BREACH, DEAR FRIENDS Daunting times are here to stay for a while – but the solar industry will not just survive, it will continue innovating. Realistically though, the pain will continue for some time and it will be diffi cult to watch industry pioneers and new entrants suffer.
The PV industry overbuilt its manufacturing capacity on the fallacies that the FiT incentive would continue relatively unchanged and that costs would improve to the point that grid parity could be achieved. The reality that the FiTs might drive too much deployment too quickly and lead to overheated markets was, by and large,
Also ignored was the reality behind low prices. In the beginning these prices were regarded as hallmarks of progress now they are transparent margin disasters
Paula Mints is founder and chief market research analyst at SPV Market Research, a partnership with Strategies Unlimited
5 reasons why you can become an Add-Energy Renewable Romania PARTNER
To register your Company as ADD ENERGY PARTNER , please access 
For other informations click to 