For the fourth quarter of 2011, CSIQ increased shipments from 355MW to 436MW sequentially. As noted in my past quarterly reviews, the company's module shipments remained strong throughout 2011 and allowed CSIQ to record annual shipments of 1323MW, up 65% from 2010. In contrast to Trina Solar (TSL) and Yingli Green Energy (YGE) which are generally regarded as higher tier suppliers than CSIQ, Canadian Solar not only met but exceeded its 2011 annual shipment targets.
Additionally, Canadian Solar has also been expanding downstream into the systems business. System kits and EPC (engineering, procurement, construction) business expanded to 11% of the company's overall annual revenues in 2011. Incremental downstream systems revenue helped buffer dramatic declines in module ASPs and as a result helped increase CSIQ's annual revenues by almost 26% vs. 10% for Trina Solar which mainly sells solar modules. With systems revenue generating gross margin in the mid/high teens percentile, consolidated gross margin also incrementally benefited from higher downstream systems contribution.
The majority of Canadian Solar's 2011 $90.8m annual loss were non-operational in nature. Net foreign exchange losses of $45.8m throughout 2011 comprised roughly half of the company's losses while another $40.6m net portion resulted from inventory and prepayment provisions. With much of the company's gross margin under pressure from the lagging effects of blending inventory costs lower towards real time procurement levels, Canadian Solar was actually profitable throughout 2011 on an operational basis when only real time procurement and sales prices were considered.
CSIQ's Q4 2011 loss was higher than expected due to several items. First the company took a $18.5m charge relating to a termination of a long term supply agreement. This was partially offset by a $14m gain from a reversal for a previously taken provision for other long term supply contract. CSIQ also reported an unusually high net foreign exchange loss of $16.5m in the fourth quarter. In addition, the company paid an unexpected high tax of $16.3m in the quarter despite operational losses due to a regional differential in tax treatment for the company's subsidiaries.
Factoring out these non-operational items, Canadian Solar lost $22.3m on an adjusted level in the fourth quarter. This compares to my $12m revised operational loss estimate. The difference was mainly due to a slightly lower module ASP of $1.02/watt vs. my $1.05/watt estimate combined with slightly higher than my estimated module unit cost of $0.95/watt. As a result, CSIQ's adjusted Q4 gross margin was 5.8% vs. my original 8.6% estimate.
As warned in my previous articles regarding Canadian Solar, the company has often stated costs which were not consistent with posted results. While blended module unit costs estimated at approximately $0.97/watt in Q4 is only slightly higher than my $0.95/watt estimate, it would have been much lower solely based on management's stated cost metrics. In a January update, Canadian Solar listed real time module unit costs of $0.76/watt. Had analysts used this figure to calculate a blended number, the result would have been even much lower than my $0.95/watt estimate.
In fact in its Q4 earnings conference call, Chairman and CEO Dr. Shawn Qu stated blended unit costs were $0.85/watt before inventory write downs which led several analysts to ask the company to reconcile the cost difference from stated and actual posted results. As a result of constant miscommunications on the part of CSIQ's management, investors should base forward expectations on estimates derived from known market factors rather than statements made by the company. These inconsistencies often make Canadian Solar appear more efficient than reality and should make investors skeptical of targets such as a $0.55-0.60/watt all in module unit cost quoted in its Q4 earnings conference call. While this target may be achieved in the longer term, it is far below what any peer with already proven lower cost structure than CSIQ have suggested.
Nevertheless, based on current market pricing Canadian Solar is in a relatively advantageous position. Recent pricing trends indicate the silicon wafer vertical is abnormally discounted, and most of CSIQ's procurement is at the wafer level. While perhaps not long term sustainable, recent spot market pricing benefits Canadian Solar relative to fully integrated firms such as Trina Solar or Yingli Green Energy. While CSIQ's "flexible integrated" model has produced inconsistent results over time, periods of dislocated market pricing have benefited the company relative to higher integrated peers.
For the first quarter of 2012, Canadian Solar once again guided for relatively strong module shipments of 340-350MW in a seasonally weak quarter for the overall industry. Combined with favorable short term procurement costs, gross margin should expand from Q4 2011's adjusted 5.8%. The company's official gross margin guidance is 5-8% for Q1 2012.
Unlike other peers, my estimates for Canadian Solar rely on derived figures based on market factors and posted results rather than comments from the company's management. Again, just based on management's comments, such as blended unit costs below $0.80/watt in the first quarter, gross margin would be significantly higher than its official 5-8% guidance. An estimate for Canadian Solar's Q1 2012 has been compiled below and mostly represent operational figures which excludes any unannounced non-operational gains or charges. While additional charges are less likely in the first quarter after annual "book cleaning" in the fourth quarter, non-operational items are always a possibility in the rapidly evolving solar industry.
Canadian Solar Q1 2012 Earnings Estimate:
- Revenues: $318m module + $50m system = $368m
- Shipments: 345MW module
- ASPs: $0.92/watt
- Blended Unit Costs: $0.86/watt
- Gross Profit: $21m module + $9m system = $30m
- Gross Margin: $30m / $368 = 8.2%
- Operating Expenses: $44m
- Net Interest Expense: $10m
- Net Foreign Exchange Gain: $4m
- Tax: $0
- Net Loss: $20m
- Diluted Share Count: 43.2m
- EPS: -0.46
Although a small $4m net foreign exchange gain is included in this estimate, actual results have often differed materially especially if the company does not keep a constant hedging practice. Canadian Solar's $45.8m net foreign exchange loss for 2011 was uncharacteristically high since the company has often managed exchange risks well in the past. The higher than expected loss was most likely due to inconsistent hedging practices which left the company over hedged in the first half and under hedged in the second half of 2011. The gain estimated for the first quarter assumes a constant hedge vs. the prior quarter.
This estimate also assumes zero taxes. Normally operational losses result in tax benefits, but Canadian Solar's global operating structure may cause regional tax differences despite overall consolidated results. This was especially evident in Q4 2011 when $16.3m of taxes were paid despite a large operational loss.
Despite an expected first quarter loss, 2012 could be a profitable year for Canadian Solar based on its annual guidance. Assuming a very conservative $0.08/watt gross profit on the median 1.8-2.0GW 2012 annual shipment guidance, CSIQ should generate over $150m in gross profits from module sales. Additionally, the company expects its systems businesses to contribute 25% of overall revenues, up from 11% in 2011. With systems gross margin likely to be fairly robust in the mid/high teens percentile, another $80-100m of gross profit could be generated.
At a current quarterly expense run rate of about $55m, or $220m annually assuming expenses are kept in check, Canadian Solar should earn a small annual profit from the estimated $230-250m in gross profit described above. If the company's cost targets are realized and module ASPs stabilize for high tier suppliers, annualized per watt gross profit level for module sales could easily expand well above $0.10/watt. Although it is too early to make forward quarterly estimates in a rapidly fluid solar market, with much of the industry's business weighted in the second half Canadian Solar should turn reasonably profitable in Q3 2012.
While unbalanced revenue recognition from solar project sales would alter the timing and magnitude of quarterly profits throughout 2012, Canadian Solar appears to have emerged as not only a survivor but a major player in the photovoltaic solar industry. The company's relatively higher exposure to lucrative markets in Canada and Japan could also beneficially differentiate its consolidated ASPs vs. peers at least in the near term. While the market as a whole may correctly price most of the industry for failure, those positioned to survive if not thrive after the industry shakeout appear to be extremely good long term values.
2011 was an extremely tough year for the solar industry as across the board declines in average selling prices ("ASP") caused losses for most companies while in more extreme situations bankruptcies and corporate shut downs. Although Canadian Solar (CSIQ) lost $90.8m last year, it could be claimed as a very successful year for the company. When the industry's ongoing brutal consolidation cycle may ultimately force the vast majority of participants out, survival in itself is victory. Canadian Solar not only survived 2011, but took significant market share gains as well.
Additional disclosure: No position in CSIQ.