"We have thrown in the towel on solar…Not that solar energy isn't a viable energy source, but we worked at it for 35 years, and we really never made money,"
I am short shares of SunPower (NASDAQ:SPWR). This is not a recommendation to buy or sell any security.
Driven by indications of bottoming polysilicon pricing, reports of new financing outlets, and a hefty post-State of the Union rally, many solar stocks are up meaningfully since the start of the year. SunPower, however, has dramatically outperformed its peers. Up 200% over the past three months, I believe SunPower's gains will be largely unwound over the next few months.
My thesis is not predicated on the idea that solar installations will decline. In the US, residential installations will continue to grow at a meaningful clip and build-out of the existing utility scale pipeline will give the appearance that the market is indeed growing at a nice rate. Abroad, Asian demand will more than offset the unit declines in Europe. Rather, my thesis is that even in the face of strong solar installation growth, SunPower's equity will not benefit.
The Short Case
SunPower is a vertically integrated manufacturer and developer of solar power systems. Depending on the quarter 70-80% of revenue comes from the US, while 10% comes from Europe/Middle East/Africa (EMEA) regions and 10% comes from Asia. In the US, SunPower's operations are best divided into larger utility scale projects and smaller residential and commercial projects.
The combination of panel costs that are roughly 50% higher than the industry average and +$360MM/yr in SG&A overhead and R&D makes SunPower's stock by far the least economic way to invest in the solar industry.
The following points summarize my short thesis:
- SunPower can't compete with First Solar's (NASDAQ:FSLR) utility scale costs. Over the past four quarters First Solar's average cost per delivered solar watt was less than $1.20. By comparison, SunPower's cost per delivered watt for the past four quarters is over $3.15. Adjusted for higher cost residential and commercial installations, SunPower's utility costs are likely close to $2.40/watt. Yes, that's double First Solar's costs. Considering the $0.0579/kwh rate that First Solar's Macho Springs project in New Mexico will receive, delivery of utility scale projects will likely approach $1/watt before incentives, and ~$1.40 watt after state incentives. It is very unlikely that SunPower can profitably add utility scale projects in today's price environment.
- The residential solar market is competitive. There are literally thousands of competitors in the space and SunPower is at a disadvantage to SolarCity (NASDAQ:SCTY) and SunRun, the market leaders.
- Hopes tied to growing markets such as Japan are vastly overblown. SunPower's Japanese market share would have to increase by several fold in order to generate free cash flow.
- The importance of Total's (NYSE:TOT) majority stake is greatly overstated. International solar markets are highly competitive and favor domestic producers. It is unlikely that SunPower will derive meaningful international utility scale profits, with or without Total's influence.
With over $410MM in net debt, I struggle to find any meaningful value in the company's equity. This despite its +$1.6B market cap. At ~$12, the stock is valued at a +12x multiple of peak earnings. I believe fair value is at least 80% less.
The market has overvalued SunPower because earnings from the company's legacy utility projects are being valued as though they are sustainable. In reality they are not even recurring after this year. SunPower's current utility-scale pipeline contains power purchasing agreement rates, which are well above current market rates. Not only will future utility projects be completed at substantially lower economics, but it is unlikely that there will be any significant utility-scale solar projects in the US for the next several years. Internationally, SunPower's current utility scale work is miniscule. This will not change any time soon. Penetrating markets is difficult and SunPower's high cost product will serve as a major disadvantage.
While SunPower has competed relatively well in the US residential and non-utility commercial solar market, margins in this business are quite low, there are no recurring revenue opportunities (unlike SolarCity, SunPower does not self-finance leases) and limited barriers to entry will only make the market more competitive. Recent speculation that residential solar assets might be prospects for REIT or MLP structures ignores the fact that investment in solar funds is already tax advantaged and these arrangements will likely result in less appealing investment options. However, no matter how you structure solar financing, even the most optimistic residential solar scenarios will not be able to cover the company's massive overhead.
Somehow the market has forgotten that SunPower has burnt over $500MM in the last three years ($191MM in 2012, $30MM in 2011 and $294MM in 2010). I believe that once everyone remembers that SunPower is not sustainably profitable, the share price will quickly revert to the $3-5/share range where it spent most of 2012.
Brief Industry Background
It is well documented that the solar industry has been plagued by an oversupply of cheap panels. Largely underwritten by Chinese subsidies, the solar panel market was commoditized and oversupplied in very short order. The result has been a race to see who could cut the most costs fastest and SunPower is not in the lead. Following is a list of current estimated costs per watt.
Cost Per Watt
Yingli Green Energy (NYSE:YGE)
Trina Solar (NYSE:TSL)
Canadian Solar (NASDAQ:CSIQ)
Since 2010, retail prices for solar modules have fallen by 65%. As a consequence, industry margins for solar manufacturers and integrated project developers plummeted. At best, margins for the largest players with downstream exposure like First Solar and SunPower were cut in half. For many panel producers, gross margins went deeply negative. Though reports indicate that some semblance of supply/demand equilibrium will be reached between the end of 2013 and 2015, there is currently +10GW of excess solar panel supply in the market (for context, the total installed Chinese solar market is 7GW) and production capacity exceeds demand by 80%. Furthermore, unlike the last oversupply cycle in 2009, when 130% year over year demand growth helped normalize conditions, 2013 global solar demand is estimated to grow at a meager 6% (Citi estimate). Rationalizing the market will require high-cost capacity to come offline. Suffice to say, margins for solar producers are going to stay low for a long time. With no economic profits to be earned producing solar panels, industry participants have collectively moved downstream.
The downstream solar market is the sales, engineering and construction part of the business. Not unlike any other E&C business, it's competitive and difficult to make outsized profits. The market is split into two general segments: utility-scale and residential/commercial. SunPower is active in both. Thus far, SunPower's downstream business has been limited to the US and Europe.
The problem with going downstream in the solar business is that there are almost zero barriers to entry, permitting large projects is expensive and risky and, most importantly, returns are heavily regulated. Remember, in all scenarios the end product is government subsidized electricity. A review of the market characteristics validates this view.
Utility Scale Solar
To say that the utility-scale solar segment is a highly fragmented industry is an understatement. In the US there are over 250 utility-scale projects currently in operation. These projects were built by over 100 different development firms. The largest, First Solar, developed 19 of them, giving them less than a 10% market share. And the industry is only becoming more competitive. There are currently around 270 utility-scale projects in some form of construction or late-stage permitting. These projects add another 100 firms to the list of companies capable of delivering asset intense, low return on capital government subsidized energy infrastructure. Following is a sample of the largest utility-scale solar developers by capacity market share.
US Utility Scale Market Share by Capacity
K Road Solar
* Includes both operating and pipeline projects
Now many industry observers will note that from 2006 through early 2011 SunPower was one of the market leaders in securing utility scale projects. By capacity, SunPower has nearly a 10% share of large projects currently in service. This share was achieved because prior to 2011 utilities were very eager to meet regulatory renewable requirements and they were willing to pay a lot more for renewable power. SunPower's products were highly regarded and power purchasing agreements were easily approved by regulators. The utilities were quick to act and now most utility pipelines for mandated renewable energy production have been built out. In California, for instance, utilities have actually oversubscribed their pipelines to meet the 2020 state mandate of 33% renewable energy production. With no incentive to exceed state mandates, utilities are not going to add additional projects. As pipelines have been built out and rates for power purchasing agreements have dropped, SunPower has had a much harder time competing.
Increased competition has clearly taken its toll on SunPower's utility scale business. Its share of current pipeline projects is about 3.2%, around one third that of First Solar. The fall-off in market share is largely attributable to price competition, and with $1/watt costs compared to First Solar's $0.64/watt costs, SunPower just isn't that competitive. While SunPower boasts that its modules are the most efficient solar panels in the world (or at least they were), this really doesn't help in large scale projects. Industry studies have shown that when it comes to utility-scale projects, mid-range efficiency offers the lowest installed cost. Not to mention the fact that SunPower's quoted watt costs have already been adjusted for its efficiency benefits.
On SunPower's most recent investor call it indicated that its current 1,000 MW pipeline of utility scale projects should generate $3.5B in revenue and $1B in gross profits over the next four years. Considering that First Solar is a much lower-cost provider (though not the lowest) and its consensus gross margins are estimated to be 24% in 2013, I find this forecast very hard to believe. However, even at a 28% gross margin, delivery of its pipeline will generate a very small amount of cash for the company. In 2013 SunPower expects to recognize roughly $1.4-1.6B of utility scale revenue (46% of its total pipeline) and yet it will generate between $100MM and $200MM in free cash flow. Looking out to 2014, utility scale revenue will not likely exceed $1B. At 28% gross margins, the $400-600MM revenue decrease will wipe out all of the company's 2013 free cash flow.
In sum, the economics of this business going forward are dismal. Rates for power purchasing agreements have been cut by more than half and the average price per installed watt for utility scale projects is declining by +30% a year. Considering that SunPower's optimistic 28% gross margin forecast is for projects with two to three year old power purchasing agreements, it is very unlikely that it could profitably develop in the current market. That is, if there were a market to be developed.
US Residential Solar Market
Recently, much attention has been given to the growing residential and commercial solar market. In 2012 approximately 428MW of residential solar capacity was installed. To first size up the opportunity, the current average cost per installed watt for a residential project is around $5 (and falling about 15% a year), so the total 2012 market was around $2.14B at current prices. Unlike its residential solar peers, SunPower does not do the installations themselves. Instead, its dealer network provides on the ground support. Since labor and other soft costs comprise 40-50% of total system costs, SunPower's addressable 2012 revenue market was something slightly above $1B. Though the company does not disclose its residential margins, the general belief is that, net of dealer markups, they are around 10% (for both cash sale and lease financing options). Stepping back a moment, this means that SunPower's total potential gross profit for the entire 2012 US residential solar market was between $100MM and $130MM, which is about a third of SunPower's annual SG&A.
The residential solar market is incredibly competitive. As a reference I'd encourage you to check out the California Solar Initiative's search tool for finding residential solar contractors here. If you search for a solar contractor in San Diego you will find 269 to choose from, with posted average costs per watt as low as $3.73. Choices and easily compared prices are great for the consumer; not so much for the service providers.
Determining a market share breakdown for residential solar installations is difficult. There are well funded new entrants coming into the market every day. In 2011 SolarCity was the market leader with approximately 14% share. Its 2012 market share is estimated at almost 19%. In Q4 SolarCity deployed 30 MW of residential solar capacity and SunPower deployed 24 MW. It seems conservative (from a short perspective) to estimate that SunPower's current share of residential installation is between 15% and 20%. Whatever the number, this share is likely falling.
In 2013, SolarCity expects to deploy 190MW of residential solar capacity, more than double its 2012 output. Reading through its 2013 guidance (flat year over year revenue with 60-70% coming from utility scale projects), SunPower clearly has much more modest expectations. Balancing available revenue inputs, SunPower's forecast for residential growth is not much more than 20% (in dollar terms). As SolarCity leverages its fixed costs (unlike SunPower, SolarCity does its own residential installations and raises most of its lease financing internally), its competitive position will only strengthen.
The residential solar installation business is more about being the best fundraiser than installing most/best panels. Installing solar panels is attractive because by selecting a 20 year lease option, homeowners can make the decision to install a solar system for zero dollars out of pocket. Since leases represent over 80% of the market (and likely going higher), developing the best financing model is the key to success. So far, SolarCity has raised almost $2B for residential solar leases through its investment tax credit funds. This is over four times the external financing that SunPower has raised. Based on comments from SunPower's CEO, the inability to finance residential solar development is having a real impact. On the Q4 2012 earnings call, Mr. Werner noted that, "Q4 lease demand outstripped our financing capacity." Though he was pleased to report that the company had also closed on $100MM in lease financing with US Bank, $100MM is only enough to fund 20 MW of residential solar development. If SunPower's residential business is growing at all then that won't be enough financing for one quarter of activity.
A recent study estimated that the 2016 residential solar market could be $5.6B in 2016. This projection strikes me as overly optimistic. At $3.50/watt, this would imply an annual installation capacity of 1,600 MW, almost four times the 2012 installation capacity. Considering that California will have well exceeded its current 5% net metering program by 2016 (after which point the utilities will not be required to subsidize residential solar production) and the current 30% Federal solar tax incentive is set to expire in 2016, this kind of growth forecast will be tough to meet.
For those banking on expanding existing solar programs in markets such as California and new markets in other regions, you may be disappointed. Programs established to require that utilities purchase solar production at retail prices will cost non-solar using Californians an extra $1.3B in annual electricity bills. Considering that these programs benefit the wealthy (large energy users benefit the most from installing solar), these incentives are increasingly being viewed as a tax on the poor. This dynamic is being explained to legislators as you read.
Whatever the case, even if the market reaches $5.6B/yr by 2016, at 20% SunPower market share (which will almost certainly have been meaningfully eroded) will struggle to generate much more than $50MM in gross profit. Again, we're talking about a company with a +$2B enterprise value and +$360MM/yr in overhead.
US Commercial Market
You will not likely be surprised to find that the US commercial solar market is also highly competitive. Almost all residential installers also are capable of completing small to midrange commercial projects. In 2011, SunPower had a 9% share of the US commercial installation market. SunPower does not breakout US commercial installations, but it is reasonable to assume that its share is similar to slightly lower. SunPower recently guided US commercial sales at $250MM in 2013. While the company does not disclose commercial gross margins (notice the trend of non-disclosure), balancing SunPower's 28% utility gross margin against its 18% combined gross margin would indicate that something around 10% is reasonable. At $25MM, the US commercial gross profit is minimal when compared with the company's $1.9B enterprise value.
Though the US commercial market will likely continue to grow at +20% in terms of capacity additions, there are more impediments to growth in the commercial market than there are in residential. Many states cap available rebates for projects, with some ceilings set at 25kW, and utilities limit the size of eligible projects for net metering programs. Furthermore, the growth in the size of the market needs to be balanced against the continued decline in solar pricing. From Q3 2011 to Q3 2012 installed costs for commercial solar projects declined 14.7% per watt. The math then follows that 20% capacity growth will only net 5% growth in the dollar value of the market.
SunPower's Asian business consists of selling panels and system components to large companies like Toshiba and Sharp who then rebrand the panels as their own. Though SunPower has been more successful than others in penetrating Asian markets while preserving some margin, at the end of the day SunPower is more or less an undifferentiated panel supplier. Remember, production capacity exceeds demand by 80% and the end product doesn't even have SunPower's name on it. For the first quarter of 2013 SunPower is expecting to sell 45-50 MW of solar capacity in its Asian segment. This is a 10% increase over the prior quarter. Throughout the year SunPower will likely sell between 200 and 225MW in its APAC market. At roughly $1.60/watt (based on Q4 2012 numbers), this segment should generate $320-$360MM in revenue. Based on Q4's 17% GAAP gross margin, this would be $54 - $61MM in gross profit. While the Asian solar market is expected to grow +20%/yr for the next several years, the pricing pressures present throughout the rest of the global market will likely be a headwind to top line revenue and profitability.
I think the word disaster best describes SunPower's operations in Europe. In Q4 GAAP gross margin was a terrifying -53%. Gross profit was -$48MM. Reductions in subsidies led to steep declines in solar installation activity. In Germany, Q4 installations were down 66% year over year. SunPower's extensive on the ground operations in Europe will likely be loss-making for the foreseeable future. In the first quarter of 2013, SunPower is expecting to install 45-50 MW of capacity, down from 50-70 MW in the fourth quarter. Considering that many of its costs are related to its on the ground installation infrastructure (i.e., fixed), investors should expect another quarter of significantly negative margins.
Future International Opportunities
There is a great transition occurring in the global solar landscape. Subsidy cuts in Europe have been replaced with heavy spending from Asian countries. The shift is not favorable for SunPower. In 2011, 36% of SunPower's revenue came from Europe. The region contributed $96MM in gross profit. In 2012, European revenue fell +$430MM to 18% of total revenue and gross profit declined by almost $150MM, going deeply negative. By comparison, revenue from its growing Asian segment increased $48MM and gross profit increased $4MM. In essence, the demand shift resulted in a net gross profit trade off of -$145MM. With European fiscal houses still in disarray, 2013 will likely be much of the same.
While the global solar industry is likely to grow, it is difficult to see how SunPower equity holders will be the beneficiaries. Simply selling panels into a foreign market is a difficult way to make money, particularly when you are a high cost producer. For large scale projects, breaking into international markets is very challenging. As shown by First Solar's acquisition of Solar Chile, the only option for penetrating new markets is often to buy your way in; probably a difficult way for a high cost producer to earn economic profits.
Saudi Arabia is often talked about as a potential locale for SunPower to build out large scale projects; its $109B investment plan over the next twenty years is noteworthy. However, it is equally noteworthy that there are around two dozen companies competitively bidding early stage projects and the recently issued white paper notes that projects will be evaluated 70% based on price and 30% on other factors such as local product sourcing and financial stability. Of note, panel efficiency is not noted as a factor for consideration.
Relationship with Total
The French oil giant, Total, owns 66% of SunPower's equity and has committed to providing liquidity to the company should it face financial duress. Many analysts have called this relationship a "Total put." Though Total's presence certainly mitigates the risk of the company going into bankruptcy, equity holders should not count on them providing much else. Total is not likely acquiring the remaining outstanding equity (if it wasn't a buyer at <$4/share it certainly isn't at +$12) and assigning value to its global relationships is misguided. By continuing to have the public shareholders as a minority investor, Total is effectively outsourcing the role of financial oversight to market. It's a pretty good idea when you think about it. And as for using its global footprint to win solar projects, there hasn't been any sign of this to date. I think it is highly unlikely that Total will expend much political capital to win low margin solar projects. Once the liquidity support agreement expires, it would not be surprising to see Total follow BP and exit the industry.
Valuing SunPower's equity is a challenge, both by way of how it is best approached and whether there is any to be found. I believe the business is best valued by forecasting the completion of its existing utility pipeline and making some estimates about the remainder of the company's business. Looking out over the next four years I come up with the following.
US Utility Rev
US Utility Gross Profit
$/MW - total project
% price declines
% Hard Costs
$/MW - hard costs
US Residential Revenue
US Commercial Gross Profit
US Gross Profit
APAC Gross Profit
EMEA Gross Profit
Increase in R&D
Total Operating Expenses
Cash Interest Expense
Free Cash Flow
Since the company provides so little disclosure in breaking out its numbers, I will admit that this represents only a best guess. However, based on company guidance and industry dynamics, I believe these estimates are generous. The scenario above assumes that SunPower retains market share where it likely will not and pegs a five year gross margin at 16.8%, a full 700 basis points higher than the company's two year trailing gross margin. Also, the assumption that SunPower "only" loses $68MM in Europe is likely quite hopeful. It lost $70MM during 2012, with $48MM of those losses in Q4 alone. Capital expenditures have averaged $160MM/yr for the last six years, so $80MM is quite friendly.
What we find here is that while the current pipeline will generate some moderate amount of cash flow for the next two years, cumulative cash generation for the next four years will likely be minimal. By the end of 2016 the company will have delivered on the entirety of its utility pipeline and will very likely be facing a combination of subsidy cuts, intense competition and demand maturation. If you roll forward the forecast to 2017, annual cash burn could be pushing $200MM. While there very well may be some utility-scale projects in the works at that time, I think it is safe to assume it won't be generating +$200MM/yr in cash flow. To say the business will have very limited prospects for generating cash is an understatement. With +$410MM in debt to be dealt with (fully drawn $275 revolver matures in September 2013 and convertible debt issuances mature in 2014 and 2015), attributing any value to the equity seems to be a stretch.
Disclosure: I am short SPWR. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.
Additional disclosure: This article represents best efforts to convey a fact-based opinion. My conclusions may be incorrect. This is not a recommendation to buy or sell any securities. I am short SPWR and may change my position at any time without notice.