Earnings Preview: Canadian Solar's Upcoming Q4 2010 Report

| About: Canadian Solar (CSIQ)

Canadian Solar (NASDAQ: CSIQ), despite its namesake, closely resembles Chinese based solar companies. Although the company is headquartered in Ontario, Canada, all of its manufacturing facilities are in China. Many of the company's executives-- including its founder, Chairman and CEO, Dr. Shawn Qu-- are also Chinese. I raise this point as many people often confuse CSIQ for a western solar company, when it more closely resembles larger Chinese based solar peers.

It is also important to point out that Canadian Solar was involved in an SEC investigation last year which ultimately led to its CFO resignation shortly thereafter. Canadian Solar was too aggressive in its accounting for fiscal year 2009, and as a result did not receive its auditor’s approval for the company's annual report filing. Some liabilities were understated, while some revenues were overstated. A complete resolution was detailed in a following earnings report which, in short, required CSIQ to take several balance sheet charges as well as defer some revenue towards future quarters. While this case has already been resolved with both the SEC and its auditors, it’s important to recognize this event did occur when discussing any of the company’s forward looking business prospects.

With these accounting issues behind them, Canadian Solar reported strong earnings for the third quarter 2010. Many operating metrics improved sequentially, which was consistent with what industry peers also reported. Revenues increased to 377.2m, up almost 15% from the 328.7m reported in the second quarter. Shipments also rose sequentially from 181.2mw to 200.4mw. Faster quarterly revenue growth was the result of new systems kit sales on top of the company's core module business. Despite a proportionately large net foreign exchange loss of 11.7m, CSIQ was still able to increase net income to 20.3m, or .47 in earnings per share.

In prior years, Canadian Solar relied on an “inverted pyramid” integration strategy. Instead of equal capacity at all verticals, as seen with fully integrated peers, CSIQ typically had more downstream module capacity than upstream wafer and cell capacity. Cell capacity was often less than half its module capacity while wafer capacity represented even lower percentages of the company's overall capacity. With this strategy, the company was able to leverage lower capital expenditure for higher market share growth at the expense of gross margin.

Another consequence of Canadian Solar’s flexible integration was that it caused Wall Street analysts to grossly misinterpret the company’s earnings. The level of CSIQ’s profitability was a direct link between the pricing spreads for each vertical. In the past analysts often normalized the company’s earnings when pricing spreads reached extreme levels. This caused forward looking estimates to either be too low or too high after pricing spreads between verticals returned to more normal levels. As a result, CSIQ often beat or missed earnings estimates by wide margins.

Like many other solar peers, Canadian Solar has also been in the process of increasing its capacity to higher integration levels. Internal cell production-- which typically represented about 50% of CSIQ’s shipments, increased to 65% in the third quarter of 2010. The company also guided this ratio to increase to roughly 75% for the fourth quarter and over 80% for 2011. Although CSIQ does not plan to increase wafer capacity to higher percentages, it does have a large wafer contract with GCL-Poly. With higher levels of integration and higher volumes of contracted wafer supplies, Canadian Solar’s earnings should be more stable and predictable, while less prone to market forces as seen in the past.

There are a few factors to consider when analyzing Canadian Solar's Q4 2010 earnings which will be released this week. Officially the company guided for shipments of 220-230mw with gross margin between 18-19%. Average selling prices were also guided slightly higher than the prior quarter. As evident with many solar peers, CSIQ has had a history of under promising and over delivering, excluding the accounting fiasco highlighted above. Given the “sold out” market condition reported by many peers, it is very likely CSIQ will report shipments at the higher end of its estimate range, if not exceed them. The company’s new systems kit division will also likely add to revenues although no official guidance was given for this business segment.

Despite CSIQ’s near full dependence on the spot market for polysilicon procurement, any virgin polysilicon purchased even at high spot levels should still result in the lowest unit costs for the company, due to full vertical integration at that procurement level. Canadian Solar’s main material procurement should still be wafers, which the company sources from higher tier wafer suppliers at contracted pricing levels. With low margin cell outsourced module manufacturing falling below 25% for the fourth quarter from 35% in the prior quarter, it should be easy for the company to expand gross margins sequentially.

With these main factors into consideration, an estimate of Canadian Solar’s fourth quarter 2010 earnings has been compiled:

Revenues: 442m
Shipments: 230mw module, 8mw system kits
Asps: 1.85/watt module
Unit Costs: 40 x 1.35 = 54m, 140 x 1.45 = 203, 50 x 1.75 = 88m
Gross Profit: 442m - (345m module) - (15m system) = 82m
Gross Margin: 88m / 442m = 18.6%
Operating Expenses: 30m
Net Interest Expense: 3m
Forex Gain: 2m
Tax: 9m
Net Income: 42m
Diluted Share Count: 44m
EPS: .96

Although estimates above are largely linked to CSIQ’s core module business, other variables could also affect the company's final results. The first is foreign exchange translations, in which a 2m gain is estimated above. Although the euro lost approximately 2% vs. the USD during the quarter-- which has typically resulted in foreign exchange losses for Chinese based solar companies, Canadian Solar’s euro exposure was only 50% of its revenues. In addition, the company was over hedged as a result of forward hedging. Assuming its hedging policy remained constant, the company should see a small gain for the fourth quarter of last year. Currency translations have typically been the most unpredictable, so actual results could differ materially.

As a result of high demand seen throughout the industry during the fourth quarter of last year, it’s possible CSIQ’s shipments are above its guidance range. Given that the pricing spread between cell and module verticals was still favorable for the company, it could have outsourced higher levels of cells and thus shipped more modules. Although at low margin, this activity would incrementally add to gross profit and thus net income. If Canadian Solar posts shipments higher than the above estimate, then revenues and earnings would benefit at the expense of reported gross margins.

Lastly, minor systems kit business was reflected in the estimates above. It represents only a minor portion of CSIQ’s earnings, thus -- on an EPS level-- results shouldn’t differ materially. Although at low gross margin, the overall level of this business segment will affect the company's reported revenues.

Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.