JA Solar (JASO) finished off 2010 strong with its fourth quarter earnings report. Metrics across the board reached record levels for the company. Annual shipments reached 1.46gw, up 187% over the prior year, but fell short of Suntech Power’s (STP) industry-leading 1.57gw 2010 shipment totals. Given that a portion of STP’s total shipments may have included externally procured cells, JASO’s cell production figures last year may have been much closer to STP's than headline numbers suggest. In fact, Solarbuzz ranked JASO tied with STP for first place in terms of global cell production last year.
On a quarterly basis, JA Solar booked record shipments of 463mw in the fourth quarter of 2010, up from 418mw recorded in the third quarter. Revenues grew at a slower sequential rate to 584.3m, mainly due to higher levels of tolling business in the company’s product mix. Net income on a U.S. GAAP basis totaled 118.7m.
Excluding non-operational gains, which totaled 53.8m (Lehman loss recovery and balance sheet gains related to derivatives linked to the company’s convertible bonds), a non-core disposal loss of 2.9m, and dilution linked to its convertible bond accounting, JA Solar posted a 0.41 Q4 2010 EPS, which missed street consensus of 0.48 in EPS. On a U.S. GAAP basis, the company's 0.59 EPS also missed my estimates of 0.70 in EPS, which did include these non-operational items. The main reason for JASO's earnings miss was linked to lower-than-expected gross margin. As a result, company shares tumbled 10% in the subsequent trading day.
Behind headline reactions, there was a deeper rationale for JASO’s sell-off. JASO missed its own gross margin guidance of 20% when it reported its gross margin fell sequentially from 22.5% to 19.2%. While it was only a small miss, it was magnified because the company had generally been very conservative on this metric. Something very unexpected happened during the quarter, which changed the company’s profitability profile.
In JA Solar’s Q4 2010 conference call, the company noted an unexpected drop in cell pricing starting in December of last year. According to Digitimes, spot pricing for solar cells had dropped from a peak level of 1.50/watt to 1.35/watt by mid-December of 2010, to as low as 1.15/watt in the following month before stabilizing around the 1.20/watt level seen recently. The spot market is perhaps the best real-time indicator of supply and demand dynamics for an industry. The sharp and unexpected drop in spot market cell pricing may have signaled a turning point in the supply and demand curve for the cell vertical. Global cell overcapacity, which some analysts have been predicting for literally years, may have finally arrived.
However, unlike the doomsday glut scenario some industry skeptics predict, pricing within the solar industry value chain may just be returning to more normal levels. 2010 was an exceptional year for the solar industry. According to a new Solarbuzz report, global photovoltaic installations reached 18.2gw in 2010, up 139% over 2009. For much of last year, demand surpassed supply, which caused pricing at all verticals to remain firm.
In such a strong seller’s market, even fresh entrants sold out any newly installed capacity. Towards the end of 2010, after total installed capacity reached 20.5gw, according to Solarbuzz, capacity finally caught up to supply. As lower tier producers found it more difficult to place their products, spot pricing dropped to reflect the new supply and demand dynamics.
As long as global demand does not drop significantly from recent levels, the supply and demand ratio indicated by Solarbuzz should not cause top tier producers problems beyond potentially lower realized selling prices. Assuming 20.5gw supply over 18.2gw demand ratio remained constant, the first “oversupply” victims would be 2.3gw of higher cost producers, who get priced out as selling prices drop below their cost of production.
If history is a guide, then higher cost producers getting priced out first would be the cell producers outside of China and Taiwan. Among lower cost producers, lower tier capacity would be selected last in the supply chain. Typically, newer companies with less of a track record or smaller companies viewed as less bankable fall under the lower tier category.
Thus, when taking into account the recent spot market pricing for solar cells, people can visualize what may be happening. Why would prices stabilize around 1.20/watt? Most likely that was the pricing level where enough overcapacity became eliminated. From a pricing rationale, some cell producers simply couldn’t manufacture and sell at this price for a gross profit.
For example, the pricing spread between spot market wafer and cell verticals at the time cell pricing stabilized was around 0.20/watt. Wafer spot pricing remained strong around 0.95-1.00/watt at the time cell spot pricing stabilized around 1.15-1.20/watt. Cell manufacturers who were entirely dependent on spot market pricing and had processing costs above 0.20/watt could not make any gross profits at these spreads.
Many less competitive cell manufacturers still require 0.25/watt or higher costs to produce solar cells, especially given recent component prices surges such as silver paste spikes noted by Canadian Solar (CSIQ) recently. Instead of operating for greater losses, uncompetitive capacity would simply get shut down. Energy Conversion Devices (ENER) and Evergreen Solar (ESLR), both U.S.-based solar companies, recently announced capacity shutdowns.
JA Solar operates at the opposite end of the spectrum, along with other tier one producers. In terms of installed cell capacity, JASO is currently the largest in the world. Larger scale allows JASO to produce cells at among the lowest cost. It also has a longer operating history than many other cell producers, especially given much of the new cell capacity installed last year were more opportunistic than strategic in nature.
As a result, JASO has accumulated a longer history of profitable operations, making financing for the company much easier than for smaller or newer producers. A combination of these factors makes its products more bankable and thus more attractive to higher tier customers. Even if JASO’s pricing isn’t the lowest, it would most likely be able to fill out its capacity first because its bankability coupled with competitive pricing offers the best overall value proposition.
In the short term, however, JASO’s main cell vertical is under pricing pressure. Although the company is perhaps best in class in what it does, JA Solar is still primarily a single vertical cell producer. JASO is still reliant on external wafer procurement for over 80% of its production requirements. As a result, its profitability is the general pricing spread between wafer and cell verticals. Under current conditions, not only is the cell vertical under pressure due to slight overcapacity, but under capacity also exists in the wafer vertical. As described above, the current spot market spread between these two verticals is not favorable for all cell manufacturers.
JASO does not generally operate at the spot market, however. Generally most of its material procurements, such as wafers, are at long term contracted prices, which are much lower than spot rates. The company also sells at prices based on less volatile contract rates. Due to the unexpected surge in demand last year, JA Solar became increasingly exposed to spot market wafer procurement to meet its increased shipments needs above and beyond the volume of its contracted wafer supplies. With the spread between the lowest contracted wafer price and highest spot market price at up to a 20% difference, JASO’s blended wafer costs rose putting pressure on its margins.
In an attempt to realign the company’s cost structure back to more normalized levels, JASO amended its large wafer and polysilicon contract with GCL-Poly late last year. In addition, the company has been gradually increasing internal wafer production capacity, which will account for approximately 20% of the wafer requirements.
However, due to the lag time required to realign its procurement volumes, the effects will not come immediately. Wafer capacity expansion is an ongoing process, while much of the new contracted wafer supplies will not arrive until the second quarter of this year when JASO expects a doubling of volume from a key source (presumably GCL).
Until costs return to more normalized contracted levels on a blended basis, JA Solar’s margins will be under pressure, as was evident in its Q4 2010 gross margins as well as its Q1 2011 margin guidance. While the recent pricing dynamics might cause weaker competitors to shut down completely, JASO should only suffer a slight margin hit. As higher volumes of contracted wafer supplies decrease blended wafer costs starting in the second quarter, the company’s margins should start to stabilize and head back to around the 20-22% level.
As a tier one supplier that only supplies to tier one customers, it’s very unlikely JASO shipments are at risk. Demand destruction would have to be extremely severe before it reaches JASO’s capacity. In a flat or slightly higher demand environment (as many industry observers are forecasting), JA Solar’s shipments should be secure; the company should continue to take market share from higher cost producers or lower tier names, as long as it can expand capacity faster than the overall growth rate of the industry.
Looking ahead into the first quarter of 2011, earnings for the company should take a short term hit but should remain at healthy levels, especially when compared to JASO’s relative stock price. Although the company officially guided for shipments of 465-475mw, its shipments should exceed those levels. In its latest conference call, JASO stated its annual capacity reached 2gw at the end of 2010 and it was “shipping everything produced” so far during the quarter. With a history of beating its own shipment guidance often by wide margins, it’s likely the company will ship 500mw or more for the first quarter. Increased percentages of internally produced wafers should also add a buffer to gross margins during the quarter.
Based on this shipment assumption, an estimate for JASO’s first quarter earnings has been compiled. Once again it uses metrics and ratios either started directly or that could be derived from company statements. Since JASO provides very little detail on its procurements as well as selling prices, the estimate below should only be viewed as a rough outline of JA Solar’s business.
JASO Q1 2011 Estimate:
Shipments: 275mw cell, 125mw tolling, 100mw module
Asps: 1.27/watt cell, 0.43/watt tolling, 1.75/watt module
Unit Costs: 1.10/watt cell, 0.23/watt tolling, 1.50/watt module
Gross Profit: 275 x 0.17 = 47m + 125 x 0.20 = 25m + 100 x 0.26 = 26m
Incremental Gross Profit: 75 x 0.20 = 15m wafer
Gross Margin: 47m + 25m + 26m + 15m = 113m / 578m = 19.5%
Operating Expenses: 23m
Net Interest Expense: 10m
Foreign Exchange Gain: 4m
Legal Settlement Loss: 4.5m
Net Income: 67m
Share Count: 172m
As a final footnote for readers still with us, the estimates above include a few factors which could change. First is the 4.5m in non-operational losses related to a class action settlement the company announced during the quarter. It may or may not be fully realized in the first quarter and the final amount may differ from the value stated in the press release.
Secondly, a foreign exchange gain of 4m is assumed. With two weeks currently left in the quarter, it’s impossible to determine the real amount at present. The foreign exchange gain assumes currency rates as of today’s date, while assuming JASO does not materially change its currency exposure by altering its hedging practices from the prior quarter.
Lastly, any adjustments linked to derivatives on its convertible bonds have been left out completely with the exception of the same dilutive effects noted in the prior quarter repeat for the first quarter. As explained in prior JASO articles, the ultimate value for this balance sheet adjustment depends on a number of variables, but most notably how the company’s stock ends the quarter relative to its share price at the start of the quarter. At present, JASO’s share price is flat from the start of the quarter, so this adjustment should be minimal. However, with two weeks of trading left in the quarter, the final value could change dramatically.