Seeking Alpha
Small-cap, macro, value, momentum
Profile| Send Message|
( followers)  


  • Solar stocks have been violently cyclical in the past.
  • But decreasing dependence on subsidies and increasing dependence on projects will reduce this significantly.
  • A less cyclical nature warrants a higher valuation for solar stocks.
  • But problems remain, like grid problems and energy storage.

The solar panel industry is one of the most violently cyclical industries out there. Since solar panels don't have all that much to distinguish themselves from those of the competition, they are a near-commodity, with pricing expressed in dollars per watt. The cyclical nature is produced by companies having little pricing power, so quantity decisions instead of pricing are more important, as is a relentless focus on cost-cutting.

In this setup, there are incentives to expand faster than the competition in order to gain market share and economies of scale. What makes sense for the individual company results in bouts of overcapacity on the industry level, which leads to price and margin destruction, and even bankruptcies, a shake-out, mergers, and companies too weak financially to build new capacity.

We just went through such a violent downturn, and it was pretty ugly. Prices of solar panels declined by up to 80% in three years, many companies went bankrupt, and the stock prices of the surviving companies were decimated.

(click to enlarge)

As a witness to the enormous gyrations in the stock price, look at the following charts of our favorite solars:

All these stocks reached a minimum around $2 less than two years ago, and have moved to $15 for Trina Solar (NYSE:TSL), or well above $30, as in the case of JinkoSolar (NYSE:JKS) and SunPower (NASDAQ:SPWR). That's quite a ride, which we warned you about (albeit a little early).

An industry cycle that multi-folds the stock prices of leading companies suggests a potentially extremely lucrative investment strategy: wait for the following (inevitable) downturn and buy the stocks on the cheap for the next upturn. One could even short the stocks if the top of the cycle nears.

One would have to keep an eye on things like capacity utilization, polysilicon pricing, and planned capacity expansions, as well as stock technicals. Considering the multiples at stake, one only needs to be roughly right (and enough margin to sit out the moves against one's position insofar as the timing is wrong).

If stock prices can collapse from $30+ to $2 in relatively short order, it doesn't matter all that much if one shorted at $25 or $35, especially since an inevitable (and infinitely more lucrative) upturn is waiting. This sounds too good to be true, and in a certain way, it is.

Not only is timing these cycles rather difficult (but one doesn't have to be exactly right), the industry itself is changing in ways that will make the solar stocks less cyclical. We will describe why and how that is.

Relentless price declines
By far the single biggest force is the relentless price declines of solar cells, caused by a combination of:

  • Economies of scale

  • Conversion efficiency gains

  • Technological progress

The cumulative effect over time is rather spectacular:

Such are the price declines, that solar can already compete with other forms of electricity generation in numerous markets. According to a Deutsche Bank report, the fall in solar panel price to around $0.60-$0.70 per watt has:

already rendered solar power competitive "without subsidies" in Japan, South Korea, Australia, Italy, Greece, Spain, Israel, South Africa, Chile, Southern California, Hawaii and Chile - in some cases because electricity prices are ruinous. (Italy's solar is not efficient but electricity retails at $0.38 per kilowatt hour, compared with $0.15 in Germany and the UK). These regions could be joined within three years by Thailand, Mexico, Argentina, Turkey and India, among others. [The Telegraph]

In the US, projects are now being built that really are competing with other electricity pricing. In Texas, city-owned Austin Energy is about to sign a 25-year PPA with Sun Edison for 150 megawatts of solar power at "just below" 5 cents per kilowatt-hour. Apart from (a usual) Federal 30% investment tax credit, this is without subsidies. Bret Kadison, COO of Austin-based Brazos Resources noted:

"This is below the all-in cost of natural gas generation, even with low fuel prices and before factoring in commodity volatility and cost overruns."... The 5-cent price falls below Austin Energy's estimates for natural gas at 7 cents, coal at 10 cents and nuclear at 13 cents. The utility points out that it approved a 16.5-cent price for the Webberville solar plant in 2009.

These are fairly large scale projects that are also dependent on financing and permissions, but the relentless fall in solar panel (and, to a lesser extent, installation) cost isn't about to come to a halt any time soon. This means that solar panels can be used to generate electricity at competitive prices in an ever greater geographic area.

What matters perhaps even more is the cost of installation, and there can be some notable differences even amongst developed markets:

Solar panels are a global commodity. Their price is essentially the same all around the world. However, the "soft costs" of a solar power system can vary tremendously. As noted back in June 2012, German solar installations cost a little more than half what US solar installations cost. At that time, German systems were being installed for an average of $2.24 per watt, while US systems were being installed for an average of $4.44 per watt. Now, US systems are probably down to about $4.00 per watt, but German systems are down to about $2.00 per watt.

The good news is, people have studied this and we have a pretty clear indication of where the costs differ.

Worldwide market
Much of the first phase of solar energy deployment was dependent upon the European (most notably, the German) market and generous government subsidies. But we're now on the threshold of a new era, brought about by the price declines, in which solar can increasingly compete with other sources of electricity production and is decreasingly dependent upon subsidies.

While the first boom was based on subsidies (that German feed-in tariff), and limited in geography, the coming boom is much more broad-based and decreasingly dependent upon subsidies. The potential market for solar was always the largest in developing countries, where sunshine is usually more plentiful and electricity tariffs can be relatively high.

And this is exactly what is starting to happen. Places like India, South Africa, the Middle East, and Latin America are starting to embark on a solar expansion that could last for many years, and parts of Asia (minus Japan and China) are not far behind.

But the immediate future in solar is likely to be dominated by China and Japan, and possibly the US. These are the top three markets in the world right now. The US, because it has pioneered new forms of financing and leasing that spread the up-front cost out and make solar much more accessible, and for distributed and utility-scale projects.

Japan, because of its ballooning fossil fuel import bill, the consequence of its nuclear disaster, and a devaluing currency. China, the biggest market of them all (the target is 14 GW of new installations this year), because of its disastrous pollution problem and the enormous overcapacity that, until recently, existed in its domestic solar sector.

This new expansion phase could be much longer-lasting, as it's much broader-based and less decreasingly dependent on subsidies. That doesn't necessarily imply that it will be any less cyclical as the previous expansion phase, as it depends whether companies together build new capacity at a much faster pace than the growth in demand. There are some reasons to be mildly optimistic, though:

  • Many solar companies still have weak balance sheets, and are not likely to embark on overly-ambitious plans.

  • The Chinese state has been orchestrating an orderly rationalization of the Chinese solar sector, which matters, as it dominates world production.

  • With decreasing dependence on subsidies, the likelihood of boom and bust markets (like Spain) due to tariff changes are decreasing as well.

Even an escalation in trade tensions between the US and China over solar panels might not be that serious:

For example, any demand downturn this year in the United States, arising from the current investigation by the US International Trade Commission against China and Taiwan, may prompt the Chinese government to increase domestic demand, in order to absorb excess supply from its local manufacturers and enable them to maintain high factory utilization rates. [Solarbuzz]

And the tensions between the EU and China over similar matters have been settled, as explained by SA contributor Doug Yong.

The main reason why solar stocks will be less cyclical is because they increasingly venture downstream into electricity-generating projects themselves (the price, and thereby revenue and earnings, fluctuates much less), rather than just selling panels.

There are two types of projects, distributed and utility-scale projects. The latter are big arrays of panels on an industrial scale comparable to other power plants.

Distributed projects are smaller and often more disjointed-scale projects producing electricity for local use and/or connecting dispersed electricity generation to the grid. There is a bit of a boom of these projects in the US, but the practice is spreading elsewhere.

The important thing to take away is that the revenues from such projects are not nearly as cyclical as selling solar panels, because they are governed by electricity rates and long-term contracts. Another thing to realize is that these projects generally generate substantially higher margins. Here is Chen Kangping, CEO of JinkoSolar during their latest CC:

We are extremely well-positioned to capitalize on these opportunities and directly benefit our project business which will yield gross and net margins at 60% and 30%, respectively.

The combination of much more stable demand and significantly higher margins is a real boon for those solar companies that are embarking on these projects, so much so that, according to Credit Suisse:

Project development growth has the potential to increase valuations for China solar companies in our coverage. As investor interest rises in solar project assets in China, companies like Trina Solar, JinkoSolar, Canadian Solar (NASDAQ:CSIQ), JA Solar (NASDAQ:JASO), and Yingli Green Energy (NYSE:YGE) should be able to finance construction costs and sell projects more readily, and/or eventually transfer projects into growth IPPs. [Barrons]

For instance, JinkoSolar started projects from zero to having a pipeline of more than 700MW today, which is already a substantial part of their yearly production (1,9GW in 2013), although not all of that 700MW will be completed in 2014. Here is Trina Solar from its Q4 CC:

For the full year of 2014, we expect our total PV module shipment between 3.6 gigawatts and 3.8 gigawatts, of which 400 megawatt to 500 megawatt of PV modules will be shipped to its own downstream projects. This would represent an increase of 39.5% to 47.3% respectively from 2013.

And the pipeline is far greater, this is just in China (they are active in the US and Europe too):

In the fourth quarter, we signed a strategic framework agreement for a 1 gigawatt project in Xinjiang, China. We expect all necessary approvals to be obtained and the construction of the first and second phases to be completed by the end of this year. In addition, we have a number of projects in the pipeline in various regions in China, most of which are currently under negotiation or in the early stage of development. We expect to see more project construction and connection to the grid as it moves into the second half of 2014.

In their own words, "2014 will be the year that we will see Trina to take off in a much faster and aggressive way to have downstream projects." [Teresa Tan, CFO, Q4 CC]

Or take SunPower, also from their latest CC:

In North America, construction on the 579 megawatt Solar Star project for MidAmerican remains on track and we reach the key milestone during the quarter as the first 57 megawatts of the project were connected to the grid. We are also pleased to announce that our CVSR project for energy is now in full commercial operation with an AC rating of 250 megawatts, CVSR is one of the worlds largest solar power plants yet completed.

In EMEA we won an 86-megawtt power plant project in the most recent highly competitive South African tender. We also saw a 10 megawatt project in Israel that was sold after completion of construction and financing, which maximize project, profit, model that we will utilize for other booked projects where the economics are compelling.

In Asia Pacific, we added another 20 megawatts of power plant bookings in Japan last quarter, which brings a total Japanese power plant bookings to more than a 110 megawatts. As I mentioned above our cooperation with Total focused on international power plant is working very well, having generated about 300 megawatts of bookings in the past year alone. We believe that together Total and SunPower are very well positioned in this market and expect to close and announced additional projects this year.

And this isn't even including their distributed projects in which, due to the high conversion efficiency of their panels, SunPower has a competitive advantage. Trefis is optimistic here:

We believe that the distributed solar business could provide a lot of growth potential for SunPower going forward considering the improving economics of rooftop solar, the availability of better financing options and the possibility of higher retail electricity prices which could make solar installations more compelling for customers of electric utilities. SunPower has a competitive advantage in the distributed solar market since its monocrystalline panels have a higher conversion efficiency and a smaller size (system capacity being equal), making them ideal for rooftops and space-constrained commercial installations. [Trefis]

Much of this is leased business. There is an excellent article by SA contributor Marsha Robe:

Total SA is the co-owner of this project, and thus SunPower was the preferred choice for project developer. SunPower has a potential pipeline of 6-gigawatt, or GW, out of which nearly one-third are in collaboration with Total.

Yingli Green declined to give guidance in their Q3 CC, but on Chinese projects, they had this to say:

The addition to market shipments we have to quite yet to believe some significant achievements in our downstream activities in China. In Q3 our revenue from downstream business increased by more than 300% from Q2. Showing great potential in this segment. Currently the construction of our solar projects is on track. We have completed approximately 80 megawatts of projects in total, which will count for two-thirds of our goals of our year goal of 120 to 130 megawatts clients for this year. For the remaining power plants, we have finished more than half of the construction and we are on track to complete them by the end of 2013.

Canadian Solar's Q4 CC noted the following:

Our backlog of late-stage solar project and EPC service contracts in Canada, now totaled around 477 megawatt. This represent an estimated revenue opportunity for over US$1.7 billion. Once the projects are built and connected to the grid 2014 and 2015... In addition, our early stage solar project pipeline, now totaled over 3.2 gigawatt. We expect a good portion of these to turn into a successful project over the next two to three years.

That's massive. However, from their Q4 PR:

At the end of January of 2014, the Company had a pipeline of late stage utility-scale solar projects totaling approximately 1.3 GW DC. These projects include owned and joint-venture projects as well as projects where we provide engineering, procurement and construction (EPC) services.

So it is sort of difficult to pick comparable numbers here, concepts differ from company to company. As a percentage of revenue:

The Company expects that its net revenue for 2014 will be in the range of approximately $2.7 billion to $2.9 billion, with approximately 50% of revenue being derived from its total solutions business. [Q4 PR]

Projects already constitute half of their revenue. So it is with some justification they could argue that:

Clearly in 2013, we make significant progress in our business transformation from a pure module manufacturer into a provider of solar energy solutions.

We find this somewhat difficult to square with their 2014 gross margin guidance of 15%, if projects are such a big part of revenue, one would expect margins to be higher. But again, you have to realize that Canadian Solar's "Total Solution Business" isn't equivalent to downstream projects:

Back in 2009, when most major PV companies were investing heavily into silicon material, Canadian Solar made a very different strategic decision to establish the total solar solution business. We formed a global PV project business unit developing PV projects in Canada, USA, Japan and China. Our team of experts specialize in project development, evaluations, system designs, engineering, managing project coordination and organizing financing. In Addition, our team of experts ensures that your project systems are designed and perfected no matter how large or small your needs maybe. [Company website]

It's basically designing projects for others. This adds value, so it should have higher margins compared to their bread and butter panel business, but not quite as owning one's own projects. Also, building third-party projects is an inherently more cyclical business compared to building one's own projects.

In the latter case, when completed, the revenue stream is almost entirely predictable, dependent upon average sunshine hours and electricity tariffs. Third-party projects depend on end demand from third-parties, which could vary much more, although the longer lead-times make revenue streams more predictable than the panel business, which enables the company to tune supply to demand considerably better.

Yingli Green has a pipeline of of utility-scale projects of 1GW in China and 200MW outside of China, but again, this isn't all for projects managed by Yingli.

First Solar (NASDAQ:FSLR), the market leader in utility-scale projects, gets 65% of its revenue from utility-scale projects, but this is slowing down:

First Solar Inc. gets about 65 percent of its revenue from selling giant solar farms to utilities, a market that's slowing after its best customers bought all the clean energy they need. [Bloomberg]

These projects are mostly managed by utilities, not First Solar, so this percentage isn't relevant to their own electricity-generating projects. The market has shifted to distributed projects, with panels build mostly on rooftops, where space limits put a premium on high-efficiency panels, not First Solar's competitive advantage. Nevertheless, it still has 3.7GW in project pipeline in the US alone.

We can summarize the above data in the following table:

Summary table

(click to enlarge)
We already stressed it several times, but not all these project business is comparable to the results. There are several business models:

  • Owned utility projects, where the solars become electricity generators themselves, connecting solar farms to the grid

  • Partly-owned utility projects, joint ventures and the like

  • Building utility projects for third-parties

  • Distributed solar projects

While all these projects confer certain advantages, in terms of benefits, projects that generate electricity to end-consumers provide the most, as these:

  • Generate stable, recurring, non-cyclical earnings

  • Often enjoy significantly higher margins (compared to selling panels)

That doesn't mean that building projects for third-parties isn't profitable, it is, and often significantly more than just selling panels, as more value is added. But while project visibility enable companies to respond to a boom or a decline, it doesn't entirely take away the cyclical nature of the solar business.

While the increasing move into projects and electricity generation will provide the solar panel producers with a degree of insulation and higher margins, this doesn't mean it's plain sailing. There are two big problems remaining, both stemming from the intermittent nature of solar energy:

  • The grid

  • Storage

The grid
The intermittent nature of solar (and other alternatives like wind) can produce problems for the grid. The main problem is one we see in Germany, where days with lots of sunshine (and/or wind) produce a boost in alternative energy. Apart from possible direct grid problems (for which there are technical solutions, but these cost considerable sums of money), this also has the potential to drive out other suppliers to the grid (and/or drive prices down).

Basically, life becomes more difficult and potentially less profitable for the backbone suppliers, as the capacity is still needed, but it's used less and possibly at lower rates. Alternative energy has the potential to make traditional electricity production a lot less profitable.

Ideally, the way to deal with the intermittency problem is to develop the capacity to store energy from alternatives like solar. The problem is, battery technology needs to get cheaper here as well. One company that has made the most waves in this department recently is undoubtedly Tesla (NASDAQ:TSLA), whose batteries:

have declined from of $500/KWh in 2008 to $250/KWh for the Model S to potentially $125/KWh at the gigafactory. [Business Insider]

This is, of course, an ongoing development, as there is something of a 'Moore's Law' even in batteries:

Tweet Ramez Naam, author of The Infinite Resource: The Power of Ideas on a Finite Planet, recently explained that lithium-ion batteries have a fifteen year history of exponential price reduction. Between 1991 and 2005, the capacity that could be bought with $100 went up by a factor of 11. The trend continues through to the present day. [Nextbigfuture]

And there might be other innovations on the way:

A new nanotechnology that doubles the life of smartphone, laptop and electric-vehicle batteries even after being charged and discharged more than 1,000 times has been developed by researchers at the University of Limerick. The breakthrough means the research team could be tapping into a market estimated to be worth US$53.7bn by 2020. [Siliconrepublic]

These are far from the only ones (see here or here). There are already projects combining solar and storage:

Sumitomo Corp. (8053) said it installed a power storage system using recycled electric-vehicle batteries near a solar power station in the western prefecture of Osaka. The 600-kilowatt system is the world's first large-scale power storage system from used EV batteries, the Tokyo-based company said in a statement today. The device uses 16 used batteries. [Bloomberg]

But solar on its own is still mostly dependent upon subsidies, add storage and the situation isn't competitive yet. So there remains lots of work to be done on the storage part especially.


The solar energy market has been growing for a long time, but this has been accompanied by boom-bust cycles for solar producers. While the smart (or lucky) investor could have made a killing on these, as stock prices gyrated manifold in short periods of time, the future looks considerably more stable and less volatile.

The inexorable move towards grid parity and the increasing incursion of main solar companies into projects will be the main drivers of this. The latter, especially, is very interesting for the investor, as projects not only provide solar companies with more stable and/or more visible earnings, they usually generate considerably higher margins as well.

Source: Solar Projects Provide Stability, Higher Margins