One of the most common misconceptions about the solar industry is that declines in average selling prices (“ASP”) are harmful to the industry’s growth and the profitability of photovoltaic companies. Of course, many less competitive and higher cost producers of solar products have been severely hurt as ASPs declined. Some companies may have been formed only to gorge on high levels of subsidies with no longer term strategy of competing once subsidies were reduced beyond a certain level.
As solar panels began steep declines in ASPs post-financial crisis of 2008, many uncompetitive solar companies have curtailed operations or ceased to exist entirely. It would be a fair statement that the vast majority of photovoltaic companies have been unsuccessful over the past several years. However, the few companies with operating metrics capable of succeeding even in a lower pricing environment have thrived at the expense of less competitive peers.
In 2008, photovoltaic module ASPs reached a peak as global demand sparked by new subsidized markets such as Spain far surpassed the industry’s capacity. Module pricing was as high as 5.00/watt for established brands, while even second tier Chinese brands reached as high 4.50/watt. In the following year, module ASPs declined by as much as 60% to under 2.00/watt.
It is important to understand that pricing did not decline because of a lack in demand. In fact, the global photovoltaic market increased by 26% from 5.95gw in 2008 to 7.5gw in 2009 despite a general global economic downturn. Module ASPs declined by another 10-20% on average for most companies in the following year, but global demand soared by 143% from 7.5gw to 18.2gw in 2010. Price elasticity in the solar industry was incredibly evident, as lower costs spurred increased demand even as subsidies declined.
Another argument industry pundits (including many Wall Street analysts) have made was that declines in ASPs destroy the profitability of solar companies. This generalized misconception caused many analysts to group the entire industry as a whole instead of differentiating companies within the industry. True, the vast majority of photovoltaic companies have experienced declines in profitability in a lower pricing environment. As noted, many have or are in the process of failing. However, a small group of large scale/low cost companies, both in thin film as well as crystalline manufacturers, have experienced significant profit growth.
One prime example in the crystalline portion of the industry is Trina Solar (TSL). According to the company’s annual report, Trina Solar’s annual average selling price for modules declined from 3.92/watt in 2008 to 2.10/watt in 2009, only to decline further to 1.75/watt in 2010. In other words, TSL’s ASPs declined by 55% over two fiscal years. In the same pricing environment which crippled many less competitive peers, Trina’s net income quintupled in the same period from $60.7m in 2008 to $96.2m in 2009 and further grew to $311.5m in 2010.
Yet throughout this explosive growth period, a number of analysts included Trina Solar in their warnings regarding how declines in ASPs would hurt corporate profitability for solar companies. Even recently, the same analysts who have grossly mischaracterized the industry’s growth potential, as well as the company’s earnings power, continue to warn how recent declines in selling prices would have a negative effect on profitability moving forward.
Because of the recent negative sentiment over Chinese companies as well as the solar sector, valuations for many US-listed Chinese solar companies have contracted into the mid- to even low-single digit price to earnings (“PE”) ratio. However, if industry leaders such as Trina Solar can manage earnings growth in 2011, bearish solar analysts may only be able to claim victory if stock prices decline on a flawed thesis argument.
Prices for PV modules decline simply because they're able to decline. No company in any industry will produce any product for negative gross margin, and the solar industry is not any different. Companies which cannot compete at a lower pricing environment have simply shut down operations. Thus, declines in ASPs have largely been a factor of the ability of leading producers to reduce costs and are not related to a simplistic linear extrapolation made by many analysts.
Due to price elasticity, it is even possible for companies to expand gross margin despite large declines in ASPs, as cost reductions outpace pricing degradation. For example, Trina Solar’s annual blended module unit cost declined from 3.14/watt in 2008 to 1.20/watt in 2010 which exceeded ASP declines of 55% in the same period. As a result, gross margin expanded from 19.8% to 31.5% during the corresponding years.
Similar to the post-2008 financial crisis, the rapid erosion in ASPs since the start of 2011 sparked renewed fears on solar profitability. ASP declines in the month of May were particularly significant across all four crystalline PV verticals. Again, the same group of Wall Street analysts pound the warning drums with the same arguments that have failed in the past. Once again, the price declines have and will continue to put tremendous stress on the remaining industry producers, which have relatively higher cost variables. The industry field will mostly likely continue to narrow. If history is any guide, companies in leading competitive positions will continue to gain market share as more rivals fade. Based on current metrics, it is likely that many leading companies in the industry, including Trina Solar, will experience earnings growth in 2011 over 2010 levels.
Despite Trina Solar’s module price declining by as much as 15% in 2011 based on the most recent guidance, unit cost reductions can potentially exceed ASP declines. TSL’s main cost components are virgin polysilicon and silicon wafers. In the past several quarters, supply and demand imbalances caused margin pressure as silicon-related procurement costs trended higher as selling prices trended lower. As history has shown, pricing imbalances between sector verticals are a short term phenomena, and since the start of May both polysilicion and silicon wafer prices have dropped dramatically. Once costs have fully normalized after inventory has been blended down to reflect current pricing levels, Trina Solar’s polysilicon and wafer costs could decline by 25-30% from Q1 2011 levels.
Since the majority of the pricing declines across all verticals occurred in May, second quarter results may be meaningfully impacted. While this pricing dynamic is not exclusive to Trina Solar, TSL will likely see its effects sooner than module producing peers such as Suntech Power (STP), Yingli Green Energy (YGE), Canadian Solar (CSIQ), Jinko Solar (JKS), and Hanwha Solarone (HSOL). This is because Trina had the lowest inventory level at the end of the first quarter relative to quarterly shipments. Relative to second quarter shipments, TSL carried less than a month of inventory heading into the quarter while other direct peers carried six to 10 weeks of inventory.
For Suntech Power and Yingli Green Energy, I am leaving my estimates unchanged, since both companies carried a high absolute level of inventory at the end of the first quarter and thus will unlikely be able to blend lower procurement costs low enough to make a meaningful impact in the second quarter. Since both estimates were made after the dramatic pricing declines in May, earnings metrics have mostly factored in recent ASP declines. Actual revenues may still be slightly lower as quotes for June pricing continued to decline but the magnitude should be small due to blending effects.
For Canadian Solar and Jinko Solar, I am lowering revenue expectations to $402m and $328m, respectively, from $415m and $344m, respectively. The slightly lower revenue expectations reflect lower blended module ASPs as pricing continued to decline after the date original estimates were made. However, since cost variables also declined, overall per watt margin levels should remain constant, and thus earnings per share estimates should not deviate too much from original estimates.
Since Trina Solar could have shipped higher cost inventory entirely prior to the pricing declines in May due to its low inventory level at the end of the first quarter, TSL is likely to experience beneficial effects of lower cost procurement sooner than peers detailed above. Consequently, new estimates have been made to reflect a slight drop in blended module ASPs as well as a more meaningful decline in cost variables. As a result, I now expect Trina Solar to report slightly higher operational EPS than my original estimate of 0.84 per share, excluding non-operational items such as net foreign exchange translations or other unannounced gains or charges.
Trina Solar Q2 2011 Earnings Estimate:
Unit Costs: 210 x 1.09 = $228.9m, 190 x 1.18 = $224.2m, 20 x 1.30 = $26m, 50 x 1.247 = $62.3m
Blended Unit Costs: $541.4m / 470mw = 1.152/watt
Gross Profit: $673m - (440mw x 1.166/watt = $507m) = $166m
Gross Margin: $166m / $673m = 24.7%
Operating Costs: $71m
Net Interest Expense: $6.5m
Net Income: $71m
Diluted Share Count: 79m
EPS if preferential tax rate is renewed and retroactively applied to Q1 2011: $1.00
In addition, euro vs. usd currency rates have corrected back towards first quarter ending levels. Currently the euro is up less than 1% vs. the usd compared to a 3-4% gain experienced last month. Although there are still 10 trading days left in the quarter, the current currency exchange rates between the usd, euro, and rmb should only have a marginal impact on overall net foreign exchange translations. Generally, for most US-listed Chinese solar companies including TSL, euro gains vs. the usd would result in currency losses if hedging practices remained constant quarter over quarter.
While peers listed above shouldn’t experience a net benefit from procurement cost reductions for the current second quarter due to higher inventory positions, lower blended unit costs should be realized by the third quarter, assuming current shipment expectations are met. First tier suppliers such as TSL, STP, and YGE have the highest percentage of maintaining shipment guidance. Although generally viewed as a lower tier group relative to TSL, STP, and YGE, second tier suppliers such as CSIQ, JKS, and HSOL aren’t excluded from meeting shipment targets. As evident in China Sunergy’s (CSUN) recent warning on second quarter shipments, lower tier suppliers have higher risk uncertainties than top tier peers. Given the industry is currently extremely bi-polar as more and more companies find itself on the less favorable side of the dividing line between winners and losers, investors should concentrate on higher tier names and less weighting on lower tier companies.
Based on current known metrics, it is likely all module suppliers listed above should experience gross margin improvements of 3-5% in the third quarter relative to the second quarter, which would mark Q2 as a trough quarter for 2011, as I’ve mentioned in recent articles. Combined with higher shipment volumes since the second half normally represents 60% of annual shipments for solar companies, second half earnings should accelerate from levels produced in the first half. With many US-listed Chinese solar companies trading at 2-5x trailing earnings, Wall Street may again be misinterpreting the earnings power of industry leading solar companies.