Semprius, a company that incubated with the National Renewable Energy Laboratory, is nearly ready to step out into the real world, announcing that its first manufacturing plant, in North Carolina, will officially open on Sept. 26.
Talk about going against the flow.
This is a time of retrenchment in the solar manufacturing sector, the result of years of exuberant capacity additions and slowing installation growth. But Semprius, with its concentrating photovoltaic (CPV) technology that yields a sky-high efficiency of 33.9 percent, figures it has something special that will allow it to flourish while competitors fade.
How did it get to this point?
The Department of Energy-backed NREL incubator program helped Semprius develop its concept with $3 million -- and guidance from experts -- in what’s now known as the SunShot Incubator, but private capital has hardly forsworn the company. Siemens put in the biggest chunk: $20 million in June 2011.
The state of North Carolina and Vance County chipped in $8 million in incentives to lure the company to the Tar Heel State.
Whether that all qualifies as government entities “picking winners” could be a matter of opinion; the federal contribution was a pittance, at least in dollar terms and compared with what Semprius raised privately. And the local incentives are de rigueur in the U.S. these days, in both blue states and red ones.
In any case, it will be fascinating to see how the company fares. Since it won the Siemens investment and said it would build a plant with a five-megawatt capacity, expandable to 35 megawatts, its task has become only more challenging as vanilla PV has become even cheaper.
In that environment, CPV manufacturer Amonix in July shuttered its Las Vegas plant, and just last week, GreenVolts suspended operations, having lost the investment of major backer ABB. ABB told GigaOM’s Katie Fehrenbacher the “decision reflects the uncertainty in the PV market, and the lack of transparency on when it will stabilize.”
Of course, none of those folks had the efficiency that Semprius can boast.
According to an NREL write-up of the company, Semprius’ gallium arsenide triple-junction cells are used in a system at 1,100 suns.
“Their tiny [chip] size means they occupy only one one-thousandth of the entire solar module area, reducing the module cost,” NREL said. “In addition, the use of a large number of small cells helps to distribute unwanted heat over the cell’s structure, so there’s no need for expensive thermal management hardware such as heat fins.”
Semprius is aiming its modules for use in commercial and utility-scale developments, particularly in very sunny areas where CPV in general is favored. For utility-scale applications, the company says its tracking, concentrating arrays will do a better job meeting peak power demand, which tends to come a few hours after fixed-tilt standard PV is at its peak.
When people ask me if investors are making money in cleantech, I tell them yes, but I don't mention which type of investors or the sector.
Most of the analyses of cleantech exits do not differentiate for venture-backed companies. So we conducted our own study.
In the last ten years, Cleantech.org’s Cleantech Venture-Backed M&A Exit Study shows a grand total of 27 venture-backed cleantech deals greater than $50 million.
All in all, very tough returns. A number of eight- to ten-figure fortunes made, just largely not by the investors spending the nine- and ten-figure investments.
We had data on both exit values and venture capital invested for 19 firms and revenue estimates for eight. We found a 2.78X Median Exit Value Multiple on venture capital invested. Those exit numbers include the founders' and management’s shares, so average returns to investors would be somewhat lower. We found a 2.2X Median Exit Value Multiple on Revenues.
There was $13 billion in total M&A exit value.
Not bad, until you realize that’s over a span of ten years where cleantech has seen tens of billions in investment, and we used a pretty broad definition of “venture-backed.” To get there, we included Toshiba’s Landys+Gyr, Total’s SunPower, EDP’s Horizon, and ABB’s Ventyx deals. Those are the top deals by value, and they represent 60 percent of the $13 billion. None were backed by investors you would normally think of as cleantech venture capital powerhouses (Bayard Capital, Cypress Semiconductor, Zilkha and Goldman Sachs, Vista Energy). Three of them included prior acquisitions themselves.
Excluding those and looking at only the transactions where we had both valuation and exit data, we found an even weaker $3.8 billion on $1.8 billion in venture capital, amounting to a 2.1X multiple.
Most surprising, if you looked at the list of investors in these nifty 27 exits, you’d have heard of very few of them. This is truly not your father’s venture capital sector.
The exits have a surprisingly low-tech flavor, and were carried by renewable energy project developers, ESCOs, and smart grid and solar balance-of-system manufacturers.
If we had limited this to Silicon Valley venture investors in high-tech deals, well, you’d have wondered if M&A were a four letter word.
Interesting, isn’t it?
Contact me at firstname.lastname@example.org if you have deal data you’d like to see included.
Neal is a founding partner of Jane Capital Partners LLC, a cleantech merchant bank whose clients have included the technology arms of multinational energy companies. He has served as a director of several technology companies and is chief blogger for Cleantech Blog. He is currently the Chairman of Jane Capital’s recently acquired Greenhome.com operating company, the original ecommerce ecostore, and serves on the board of American Electric Technologies, Inc. (Nasdaq: AETI). He is a cofounder of venture-backed smart grid company Smart Wire Grid, Inc., where he served as founding CFO and director, and venture-backed carbon IT startup Carbonflow, a Jane Capital spin-out in SaaS workflow for the carbon markets, where he served as founding CEO and raised the first two rounds of finance.
This article appeared originally on Neal's Cleantech Blog.
Four years ago, when Consolidated Edison was designing its commercial and industrial rebate programs, LEDs weren’t mature enough to make it into the prescriptive program. Instead, solid-state lighting was lumped in with the custom rebate program.
When the programs are redesigned next year, some LEDs will probably move out of the custom category. “In the past few years, the LEDs have come on fast,” said David Pospisil, program manager of the Con Edison Commercial & Industrial Energy Efficiency Programs.
Many companies are looking to their utilities to provide rebates for LED retrofits, which offer a huge energy savings but can still come with a significant upfront cost -- despite falling prices. Davis & Warshow, a nearly 90-year-old family business in ConEd’s territory, recently spent more than $250,000 to retrofit its Queens distribution center from fluorescent tubes to LED tubes.
The retrofit was the largest single-facility LED tube installation with the most energy efficiency lighting in watts per square foot to date, according to ConEd. Before the switch, lighting ate up nearly 65 percent of the building’s total power usage, costing nearly $50,000 per year. Now that figure is less than $20,000, about 60 percent lower.
The retrofit was possible in part to ConEd’s commercial and industrial rebate program, which gave Davis & Warshow a rebate check for $63,704. The project is expected to have a five-year payback. The company, based in Maspeth, NY, also has a rooftop solar array that puts out more than 200,000 kilowatt-hours annually, which makes the new lighting system net-zero.
ConEd is not alone in treading lightly into shifting LEDs from custom rebate programs to prescriptive programs. Utilities across the country show limited prescriptive rebate support for LED lighting, according to a report by Groom Energy and GTM Research, Enterprise LED 2012: Commercial and Industrial Market Trends, Opportunities and Leading Companies. But that is changing.
New York is second only to California in dollars spent by utilities in energy efficiency rebate incentive programs. In 2008, there was about $3.1 billion in total U.S. rebate dollars, with the money concentrated in ten states. The figures are expected to more than double in coming years, with $7.4 billion to $12.4 billion available by 2020.
As the LED market matures, utilities will aggressively start moving. “When satisfied that savings can be successfully achieved, utility program managers will typically authorize custom rebate amounts up to 50 percent of the entire cost of the project, as opposed to a prescriptive rebate for each fixture,” the report authors wrote.
ConEd falls squarely into this category. The utility, which serves New York City and some surrounding areas, is seeing a booming demand for LED lighting, but is still waiting for more standards. The report notes that other Northeast utilities, including National Grid and NStar, are some of the most progressive, offering prescriptive rebate programs for a range of LED lighting applications.
Commercial customers look to the utilities to be a trustworthy source when it comes to picking new products, which is why so many utilities are wading carefully into the world of LEDs. “There are a lot of products that are substandard, and it’s hard for customers to separate products,” said ConEd’s Pospisil. Inclusion on the Design Lights Consortium-approved product list has moved to being the key criterion for most utilities, according to the Groom report. Rather than creating a standard, DLC is a group of utilities and energy efficiency groups that approve lighting products.
While utilities don’t want to back the wrong horse, they also don’t want to miss out on the growing interest in LED retrofits. ConEd has $50 million per year to spend through the end of 2015 on C&I rebate programs total.
California, New York, Florida and Massachusetts have some of the most robust energy efficiency programs, but others, like Pennsylvania, Illinois, Arizona and Ohio, have started building new programs entirely, according to the Consortium for Energy Efficiency. North Carolina and Michigan are also increasing spending, according to Lawrence Berkeley National Laboratory.
The falling cost of LEDs is a driver for the interest on the part of customers, but it is only one piece of the puzzle, according to Jon Guerster, CEO of Groom Energy. Even if states that are new to the game aren’t putting in as much money as California, for instance, the increased utility funds will help push LEDs into the commercial and industrial mainstream.
“The answer is not just that this is cheaper,” he said. “Our customers say, ‘LEDs are here and they’re still expensive, but I have to be thinking about it.’ That’s a massive shift.”
“Then the price of silicon collapsed,” Chatila said. “As 2011 wore on and supply exceeded demand significantly, MEMC struggled.” In response, he said, “we became aggressive in restructuring the company.”
Before moving from Cypress Semiconductor (NASDAQ: CY) to MEMC in 2009, Chatila said, he visited 150 solar and semiconductor companies worldwide. “I came to the conclusion that everybody would get crushed,” he said. “But I thought it would be 2014-15. I thought I had enough time to build capacity, do innovation and grow downstream. Unfortunately, I was wrong.”
In the solar business, Chatila said, “you have to have volume, price, and costs. When you own the customer, you own the price and you own the volume.” In a commodity business, Chatila added, “if you don’t have the lowest cost in the world, you don’t win. When I walked into the company, we had lost that race.”
But, Chatila said, “we had customers and long term agreements. We had to satisfy these instead of just building capacity.”
Though they could not avoid the crisis of 2011, Chatila said, “the team and I understood that the way you win is through technology innovation upstream and focusing on brand and customer relations downstream. When we get more pipeline and service customers downstream, we profit. And if we differentiate our technology upstream, we profit.”
MEMC is a complex business, Chatila said. “The supply chain begins with polysilicon, then you make a crystal, a wafer, a cell, and then a module. Then you use other products to install and you sell the project, by financing it. And then you maintain the project. We make all the products. We don’t manufacture modules ourselves, but we have partnerships in making them. And we do development and operations and maintenance.”
The cheapest place to make polysilicon and crystals is the U.S., Chatila said, because factories are deployed in low-cost-electricity states. The cheapest place to manufacture modules and cells is China, because that part of the process is labor-intensive.
Most balance-of-system (BOS) products, except the inverter, he said, are made near the project, because transporting them adds to project costs. And the best inverter companies are German and American, Chatila said, because they have first-mover advantage.
Development, Chatila said, “is complicated. You have to be global. But decision making is always local. If you are tied to one country, one day you will wake up and there will be zero demand.”
The solar industry is trending toward midsize projects, Chatila explained, because “banks don’t like the high risk of projects above 100 megawatts and, because of that, you have to give them a lower price. And projects of less than one megawatt are not big enough to be interesting to banks and investors.”
Banks and investors, he said, “love 20- to 50-megawatt projects. They are under the radar, they are not very complicated and low risk. Banks don’t want to put a half-million dollars into a $3-million, one-megawatt deal. They want to spend a half-million dollars on a $100 million deal. And they don’t want to spend a half-million dollars on a $1 billion deal. That’s very high risk.”
MEMC recently moved its headquarters to Belmont, California, because, Chatila said, “we think California is the most interesting market in the world.” There is a business for residential and small commercial development, he said, “but you have to do it as a flow business.” California’s innovation in third-party ownership and other project finance models is facilitating such flows, while “in some places, third-party ownership is not as allowed as it should be.”
There are three elements in project finance, Chatila said. “During construction, you need money to build the project, before it is ready to be sold. When it is sold, there are two elements. One, the debt, you get from banks. The other, equity, you get from investors. We bring all that money together. That is one of the complexities of doing downstream.”
In MEMC’s Bulgaria project, he added, “The investors are from the U.S., Turkey and Saudi Arabia. The debt holders are OPIC, IFC and Italy’s Unicredit. On one project, 60 megawatts, a quarter-billion dollars, you have three investors and three debt holders, one from Italy, one international and one is U.S.”
MEMC is building around the world, Chatila said. Panel cost is “70 cents per watt, plus or minus.” The costs for building a project range “between less than $2 per watt to $3 per watt, depending on the country and the size of the project.”
From his experience in the semiconductor business, Chatila knew the Chinese would be able to make silicon modules cheaper than their competition. When they began taking market share, he said, other countries began imposing tariffs. “When you do that, you are saving 3,000 jobs. I do care about those jobs. But, since 80 percent to 90 percent of solar jobs are local, you are going to destroy 97,000 jobs, which I care about even more.”
And the tariff, he added, “is not impacting anything. When two countries go to war, guess who gains? The countries not at war. Guess where all the business is going -- to Korea, Malaysia, and Taiwan.”
On the strength of overwhelming Republican support, the House of Representatives voted today to torpedo a U.S. Department of Energy loan guarantee program that by objective analysis has fared better than it was expected to when a Republican-led Congress and a Republican president invented it.
The “No More Solyndras Act” won’t become law -- the Senate has no appetite to take it up and President Obama would surely veto it if it somehow landed on his desk.
Republicans -- who voted 223-4 for the bill -- were well aware of that, but said it nevertheless was important to send a message about where they stood.
Keystone, of course, is a reference to the pipeline that would bring oil sands petroleum down from Canada, which environmentalists oppose for a number of reasons, not least of which is that the Canadian government says [PDF] that boosting production of the stuff will account for almost all of the country’s increase in greenhouse-gas emissions up to 2020.
Solyndra? Well, surely you’ve heard about Solyndra; the Republicans have been beating on the Solyndra drum for more than a year now, since the DOE-loan-backed company failed and swallowed up more than a half-billion taxpayer dollars in the process.
Despite that DOE face-plant, an independent analysis of the loan guarantee program determined that it actually “holds less than the amount of risk envisioned by Congress when it created and funded the program.” Also, a Bloomberg Government analysis found that since nearly 90 percent of the loan program was invested in energy projects that had buyers for the power they will produce, the risk of big losses was minimal.
Democrats voted against "No More Solyndras" -- the bill, not the concept -- by a 157-22 count. “This is not serious legislation, it’s a political bill,” Henry Waxman of California said, according to Reuters. “They’ve been dancing on the grave of Solyndra for so long. Enough is enough.”