JA Solar (NASDAQ: JASO) is probably one of the more misunderstood Chinese solar companies in the sector. The company started off as a single vertical cell producer, but since the start of this year has slowly added both upstream and downstream capacity to be fully vertical at a portion of their overall capacity. For a very long time, management continued to stress they were only interested in their core business of cell production. Of the three main verticals of wafer, cell, and module production, the cell vertical is the most technologically dependent which can differentiate companies more than in other verticals that are more experienced or supply chain management dependent. In order to be successful, cell producers have to develop their technology to balance cell efficiency along with production costs. The highest efficiency cells may not have the best or the lowest production costs per piece. It’s the balance of providing the highest efficiency possible while still maintaining the lowest production costs possible that is critical to a company’s success. This is what JASO has been able to do extremely well in its short history as a company.
There are limitations on being a single vertical however. When the industry is in an up cycle, single verticals can perform disproportionately better since they can leverage their lower per watt capital costs to increase shipments. The problem arises during down cycles where margins may contract and there are not enough per watt gross margins for single verticals to maintain the same level of profitability. Fully integrated companies in contrast can cut out one or more middlemen and thus keep more of the per watt gross margins the industry at that instance offers. In addition, being an upstream provider such as JASO makes the company extremely dependent on how well their customers perform. As a result, lesser integrated companies in the sector may see higher and lower peaks and valleys in their business whereas more integrated companies may see higher stability in their business.
So the question for quite some time has been: why didn’t JASO integrate sooner? The answer(s) is a bit more complex. First, full integration costs a lot of money. During up cycles everyone can do well so no one worries about integration, but usually when the advantages of integration become more apparent during down cycles, capital is more tight since business conditions are usually at their worse. Secondly, JASO had already made numerous multi-year wafer agreements with their suppliers prior to the period before anyone even considered the merits of integration. Among the US listed Chinese solar companies, JASO has among the highest levels of contracted supplies and has made among the highest levels of prepayments for it. Even if the company had the means to integrate upstream, they could only do it to the level of their capacity not already covered by supply contracts.
While upstream integration might be capped for JASO, the company isn’t as confined with downstream module integration. The questions and answers work a bit differently for this vertical. Adding module capacity is extremely cheap and generally there are no technological barriers. The vertical becomes more a factor of logistics and good supply chain management. As a result of these factors, the module only vertical carries extremely low gross margins which can disappear completely during business cycles where there is a disconnect between pricing levels among different verticals. Many have asked, why even bother with module integration? In fact, JASO has stressed they themselves have no interest in this vertical. Not only would they have to deal with the issues mentioned, but branding your own modules takes a lot of effort and resources which may only bear good operating results after years of business. Instead, the company is only adding module capacity per customer requests.
Since the solar cells JASO produces are not the final product, customers have to convert purchased cells into modules. Some customers already have strong module branding along with low manufacturing costs. Other customers especially higher cost non-Chinese module producers end up tolling JASO’s cells for module processing. The company itself mentioned it has referred other companies to their cell customers for module processing. However as module selling prices have declined over the past couple of years, so have per watt gross margins. As a result, everyone especially higher cost producers have been seeking to cut as much costs as possible. Hence instead of taking the extra step in having module production outsourced, many have asked JASO to do it. In essence, JASO is only providing white label oem services for current customers as an add-on service. The result is added per watt gross margins albeit however small at very little additional overhead costs.
The end result is a hybrid integrated company with some upstream capacity to help reduce costs and downstream capacity to provide more value for their customers. In other words, JASO is essentially capturing higher levels of per watt gross margins with the same customer base without having to spend too much in terms of additional resource. This translates to higher overall gross margins as higher levels of this hybrid integration begin to take effect. Because the company only started this model earlier this year, the effects have not been seen in their earnings report yet. For the third quarter of 2010, we should start to see limited effects and higher progression in the following quarters.
This brings us to the quarter which already ended. We know demand has been higher than capacity for many of the key players including JASO. We know that pricing across most verticals have risen to some degree. As far as JASO is concerned since they have a high level of contracted wafer supplies, they should only see modest increases in wafer costs. Although cell pricing has also been strong, the company has also taken the more strategic approach by continuing to offer their customers preferential pricing in order to solidify their customer base. The end result is JASO will sell out whatever capacity they have while maintaining stable levels of per watt gross margins along their core cell vertical. Since JASO withholds many key metrics, estimating their earnings is a lot more tricky than for similar peers. Below is an estimate based on their guidance along with other backwardly derived metrics based on prior reports.
Shipments: 205mw, 95mw oem, 75mw module
Asps: 1.32/watt cell, .45/watt oem, 1.75/watt module
Unit Costs: 1.04/watt cell, .23/watt oem, 1.45/watt module
Gross Profit: 205 x .28 = 57.4m + 95 x .22 = 20.9m + 75 x .30 = 22.5m
Incremental Gross Profit: 30 x .15 = 4.5m wafer
Gross Margin: 57.4m + 20.9m + 22.5m + 4.5m = 105.3m / 445m = 23.6%
Operating Expenses: 21m
Net Interest Expense: 7.5m
Net Income: 64.8m
Share Count: 165m
The earnings per share listed is a core operational number which excludes a couple of non-operating items that will affect the overall US GAAP results dramatically. The first are currency translations. For Q3, the main currencies involved are the euro and the rmb. Both currencies gained against the usd which should result in gains at each level. JASO should report gains on the euro side for assets in euros such as accounts receivables. On the rmb side, the company should also report gains because they have a large usd denominated liability - their convertible bonds. Based on the foreign exchange levels ending Q3, JASO should post at least a 5m currency gain. The other item is the fair value adjustment on derivatives linked to their convertible bonds. It’s nearly impossible to estimate with any degree of accuracy the absolute impact of this line item. The main contributing elements are volatility, stock price, and credit spreads of which the stock price factor will have the most impact.
In short, the more JASO stock goes up and towards the convertible bond conversion price, the higher the value of those bonds will be on the company’s balance sheet. Since this liability increases with the stock price, JASO will post losses which could be substantial since their stock price roughly doubled during the third quarter. Based on prior reports, it would not be unreasonable to estimate this loss ranging 15-20m for the quarter, or perhaps even more. Since these two items are much harder to predict than the company’s operational business, only a rough estimation is possible. Combining these two factors will probably yield a negative .06-.12 EPS for JASO’s Q3 earnings. As a result, JASO’s US GAAP Q3 earnings might fall around the .30 EPS level instead of the .39 operational EPS listed above.
There is one last caveat however. In the most recent quarters, JASO has been extremely conservative with any guidance they have given. As a result, the company has generally beaten their shipment and margin guidance by fair degrees. If this were to occur again, their operational numbers could be higher. For instance, the company guided for shipments of 375mw, but that number could be around 400mw based on other direct and indirect information. Margins could also be higher since the company was conservative in indicating higher wafer procurement costs which was reflected in the estimates above. However, in a recent earnings report by a key supplier, they reported no meaningful increases in selling prices which JASO would have to bear.
All in all, JASO is likely to report an extremely positive quarter with the only uncertainty being the non-cash non-operational derivative loss linked to their convertible bonds. In a rational market, investors would understand this derivative loss as an accounting formality and exclude them from operational results. However this may not always be the case depending on stock market sentiment at the time of earnings. For example, JASO also reported the same derivatives loss in their Q2 report which was misunderstood and caused the stock to drop 10% after their earnings release. As always, time corrected this misinterpretation as seen by JASO’s performance after their last earnings.
Disclosure: Long JASO.