Solar
Summary:
§ This email expresses a bearish view on the solar industry, specifically with regard to manufacturers of solar photovoltaic cells
§ Although demand is growing, and over the long-term the solar sector will become very large, there is currently a massive oversupply of solar PV cells. This oversupply should become much worse in 2010 and 2011. Solar cells are essentially commodity products, and the large supply-demand imbalance should render most high-cost producers insolvent. Low-cost producers will also have trouble generating profits.
This week, I’m going to write about companies that sell solar photovoltaic cells, and how I think 50%+ may go bankrupt over the next several years. I think solar cell companies are in the fourth inning of an unmitigated disaster, and the looming supply-demand imbalance in the industry is one of the most violent I’ve ever witnessed in any commodity sector.
Unfortunately, I think that the money-making opportunities are limited for conservative investors. Indeed, when I initially decided to write about the doom facing the solar sector, I was fairly excited about shifting a material portion of my short / put / hedge book into solar companies. In the end, however, I came away a little less enthusiastic. The problem is that (a) I’m not sufficiently certain that I’m right, given the high threshold one wants to pass in order to put on aggressive positions on the short / put side, (b) if I’m wrong, I can lose a lot of money and (b) given that valuations have come down materially over the past year, it’s prudent to identify companies that are very likely to go bankrupt.
And I didn’t find any of those surefire bankruptcies, although Q-Cells comes close. In investing, few things are more important than having conviction in one’s investment theses. While I think I’m right about this one, I just don’t have enough conviction to put the trade on in large size. Basically, I’d be proven wrong if demand for solar cells soars over the next couple years beyond most analysts’ estimates, driven by large government subsidies and fast-declining module prices. I doubt that the unexpectedly large demand will materialize, and I think that by the time demand outstrips supply, most current solar cell companies will have disappeared. But it’s possible that demand accelerates quicker than I’m expecting. These caveats aside though, I wouldn’t be surprised if the stocks of the three companies discussed below go much lower, or the companies go bust altogether.
Industry Description
The solar sector is part of the emerging renewable energy sector, and over the next 20 years, the solar energy industry will grow rapidly. If even 5% of the world’s electricity is generated by solar power, the size of the industry would increase by over 10 times. By 2020, there will probably be solar cells on top of every building in New York. The government may even pass laws requiring it. But just because the solar sector will experience massive growth doesn’t mean that the current manufacturers of solar cells are poised for a promising future. Today, just about every person in the US uses broadband communication, and the growth in broadband in the past 20 years has been tremendous. But most companies that laid fiber-optic networks in the ground during the 1990s went bust. Solar is similar. Industry growth translates to increased profitability for a given company as long as demand for a company’s products is greater than the supply of those or competitors’ products. When supply outstrips demand, regardless of how fast demand is growing, high-cost suppliers go bankrupt. The demand for solar cells has been increasing at a rapid pace, and will continue to do so. But the supply of solar cells has increased even more rapidly, and it will easily overwhelm demand in 2010 and 2011, in my opinion. It already has, and we’ve seen solar stocks decline 80%+ in some cases. The pain isn’t over yet.
There are two leading technologies in the solar space – silicon-wafer based and thin film-based. Our focus will be on the silicon-wafer based cell manufacturers, which comprise more than 90% of solar cells. I think the main thin film solar cell manufacturer, First Solar, also faces a bleak future, but it has a lot of cash and some long-term contracts with customers, so I’m not focusing on First Solar in this writeup.
There are 5 broad steps to making silicon-wafer PV cells. Step 1 is to make polysilicon. Step 2 is to convert the polysilicon into ingots and wafers. Step 3 is to convert these ingots and wafers into solar cells. Step 4 is to take these cells and put them together into solar “modules”. Step 5 is to construct / install these modules into solar “systems” for customers. Different companies focus on different steps within this 5-step process. Some participate in a combination of steps, such as steps 1 and 2, or steps 2-4, or steps 4-5. The most profitable steps, historically, have been steps 1 and 3. Our focus here is on companies that focus on step 3, or the manufacturing of the cell, but many of the Step 3 companies also participate in other steps in the manufacturing process. No company operates in all 5 steps, yet.
Each step features its own supply-demand characteristics, and that supply-demand imbalance will govern whether the price of the commodity produced in each step will be priced above, below or at marginal cost. Basic economics tells us that if there’s too much supply of a given commodity, the price of that commodity will trend towards marginal cost, or below marginal cost until demand catches up to supply. Currently, there is too much supply in all the steps. There is too much polysilicon. There are too many wafers. There are too may solar cells and there are too many modules. The oversupply began in 4Q08, and has only become more exacerbated as time has gone on. Polysilicon prices have crashed from about $400/kg to about $70/kg. Marginal cost is estimated to be around $35 to $45/kg, and I’ll bet that prices will get there soon enough.
As for solar cells, below is the most important table of the solar short / put thesis. I’ve laid out the demand for PV cells, compared with the supply of PV cells, historically and projected, as provided by two sell-side solar analysts. One is from Barclays and is bullish on the industry. The other is from Hapoalim and is generally bearish on the industry. Some numbers are estimated, and the projections were composed at different dates, but both forecasts tell a similar story.
In 2006 through 2008, demand grew rapidly and outstripped supply. This can be seen by the negative numbers in the “Supply Minus Demand” rows, which simply designate PV cell shipments less demand. In 2009 through 2011, both analysts project massive oversupply. The supply-demand imbalances here are not rounding errors. The annual oversupply of photovoltaic solar cells is projected to be 30%-50% of total PV cell shipments!
That’s the crux of the bearish thesis. There’s too much PV cell capacity – way too much – and, as a result, high-cost PV cell producers are going to become deeply unprofitable. Prices for PV cells and modules will continue to drop until either supply goes away (high-cost manufacturers liquidate or otherwise shut down capacity), or until demand surges in response to the lower prices and catches up with supply. While I’m sure that demand will surge in response to the lower prices, I don’t think it’ll be fast enough to save high-cost suppliers.
Selected Companies
In choosing short / put candidates to discuss, I’ve avoided companies that are (i) low dollar-price stocks and (ii) have market caps under $1bn. Some of the solar companies that I’ve seen, particularly the ones in China, have very questionable accounting and have hemorrhaged cash since inception. But as much as I think “Solarfun” is not going to be around in 5 years (I’m serious, there’s a company called “Solarfun”. Its website features 4 children dancing and skipping around a solar cell), it’s nevertheless a $300mm market cap company with a $5.00 stock price. Different investors have different philosophies on this – I personally get queasy watching speculative smallcaps whip around.
Given those constraints, I’ve looked for companies that feature the following:
(i) high leverage, as evidenced by Debt-to-EBITDA and Debt-to-Cash-Flow ratios
(ii) high cost structures
(iii) limited systems/installation business – ie. they don’t participate in step 5 of the solar cell process that we discussed above, and therefore don’t have customer relationships that they can use to differentiate themselves from other PV cell manufacturers
(iv) low efficiencies (this refers to how technologically advanced their cells are in terms of absorbing sunlight. Higher efficiency cells absorb more sunlight than lower efficiency cells)
(v) high valuations
The finalists I’ve decided on thus far are Q-Cells SE (QCE on the Deutsche Borse); Suntech Power Holdings Co., Ltd (STP on NYSE); and SunPower Corp. (SPWR.A on Nasdaq).
Q-Cells is one of the highest-cost producers in the solar PV cell space, due to its manufacturing plants in Germany. It is transitioning its facilities to Malaysia, but the move is a bit late, and costly. Leverage has creeped up, with net debt to LTM EBITDA above 4x and likely to creep higher, as EBITDA goes lower and net debt increases throughout 2009 and 2010. Just about the only thing Q-Cells has going for it is that its next debt maturity, a €420mm convertible issuance, is not due until February 2012, and the company has ample cash and availability under its Malaysian revolver until then. But February 2012 is only 20 months away. EBITDA in the last two quarters has been €30mm in Q1 and -€40mm in Q2. Its market capitalization is around €1.2bn. Its PV cell efficiency is not particularly high, standing at around 17%, compared to 22% for Sunpower, and 15% to 17% for Asian producers. It doesn’t have material exposure to United States or Asia, which have upside potential if the US or Chinese governments introduce new feed-in tariff subsidies for solar energy. Its systems business was begun a year ago, and therefore the company somewhat falls into the category of a PV-cell-only producer with limited customer relationships.
Suntech is the largest Asian producer. Generally, I think the Asian solar cell manufacturers are among the better positioned players in the industry given their lower-cost status. But there’s several things I don’t like about Suntech. It has a lot of debt, and leverage stands north of 4x. Its debt situation is a bit murky though, because much of its borrowings are with Chinese banks, and I suspect that the Chinese banks will continue to support Suntech as leverage goes higher. Suntech has also locked itself into a material number of long-term silicon supply contracts. As background to this, from 2005 to 2008, there was a shortage of polysilicon, due to the unexpectedly fast nature of the ramp-up in solar demand beginning 4-5 years ago. The limited supply of polysilicon allowed cell manufacturers with access to polysilicon to enjoy strong margins and pricing power. Polysilicon suppliers, in turn, demanded that cell manufacturers sign long-term contracts at fixed prices, as well as make large cash prepayments to ensure future supply. Now that polysilicon prices have plummeted, these contracts may prove to be a burden to cell manufacturers that agreed to them in the 2006 to 2008 timeframe. Contracted prices are around $60 to $80 per kilogram, whereas spot polysilicon prices will probably head under $50/kg. It’s difficult to get color on which cell manufacturers are locked into the most onerous contracts, but I think Suntech is amongst them. According to one analyst, the company’s estimated next twelve-month (NTM) minimum raw material purchase obligations as a percentage of NTM sales amounted to more than 90%, while the average for its peer group is closer to 50%. As for valuation, STP’s EV / EBITDA multiple stands at more than 20x, due to plummeting EBITDA. And that figure includes more than $100mm of revenue that STP has booked to a joint venture it owns, a suspicious piece of accounting that STP may have used to inflate Q1 numbers.
Sunpower is the largest U.S. producer of silicon-wafer based solar cells. It is a best-in-class provider and features the highest cell efficiency in the industry at 22%. It also has an established systems business and savvier marketing in the US than other cellmakers. Its modules currently sell at a premium to other solar cells. It has been faring ok, compared to other solar suppliers, over the last several quarters because it has taken a lead role in the U.S. and California markets, which have been growing faster than European markets (specifically Spain, which has contracted rapidly this year). Over time, however, I think that Sunpower will be exposed to competition from China, and I think that its higher cell efficiencies, marketing prowess and customer relationships will not compensate for the fact that Chinese producers may flood the market with cheap PV cells while their cash burns are financed by government banks. I think that the increasing oversupply of PV cells in the next two years will impact all industry players, and Sunpower will continue to be impacted, via eroding margins, negative cash flow and increased debt / equity dilution. Sunpower has both an A share and a B share, which irrationally trade at materially different prices. The A share has high short interest due to arbs trying to capitalize on the difference, and the B share has a lower dollar price.
A final point: solar cell manufacturing is a technology-based industry, and it’s unclear whether the current silicon-wafer based technology will prove ultimately triumphant. There are numerous upstarts that are trying to develop cheaper, more efficient cells than those currently produced by the silicon-wafer and thin film technologies. One is Nanosolar, and it is instructive to check out the technology section of its website at http://www.nanosolar.com/technology.htm. There is a material chance that new future technologies render most current cell manufacturers irrelevant.
This email does not constitute investment advice or a recommendation of any sorts. I may buy, sell or short any of the stocks mentioned at any time. I may be wrong; it won’t be the first or last time.
As always, I look forward to your thoughts.
