A couple weeks back cellulosic ethanol hopeful Mascoma quietly pulled its IPO after it had filed to go public. It’s safe to say that the abysmal performance of the 2011 crop of biofuels IPOs, that included Solazyme, Gevo and Kior, did not help Mascoma’s chances of finding public money. Gevo’s been a particular stinker, losing over 80 percent of its value, though Kior is down almost 70 percent from its IPO pricing.

My colleague Katie Fehrenbacher has analyzed the various issues related to high levels of risk and companies like Kior. With no revenue at IPO and no significant revenue on the horizon until the company built a capital intensive production facility costing tens of millions, Kior has carried a significant level of risk for a publicly traded company.

I mention the high risk world of biofuel investing and the fact that these companies went to public markets seeking capital because two years after the 2011 class of biofuels IPOs, cleantech investors find themselves in challenges situations where it’s difficult get IPOs done. And without those investors having access to liquidity and returns, late stage companies are having harder times finding capital to push through the commercialization phase of growth.

What happened?

The IPO market all but dried up with just three cleantech IPOs in 2012 and overall cleantech VC dropped by a third. IPOs are critical for venture investors to find liquidity and produce returns, as is significant M&A activity. But IPOs that significantly underperform the market make it harder for other companies in that sector to attract VC or to go public themselves.

Of greatest concern is that as financing for cleantech gets tight, the brightest startups will struggle to find early stage capital and those companies nearing the path to commercialization will find it hard to find scaling capital.

Many of these financing issues are cleantech specific. It’s worth looking at Matthew Nordan’s analysis of the state of cleantech investing but some of the key points he makes are:

• 1) The current value of cleantech funds is about 90 percent of what LPs put into those cleantech specific funds, versus about 123 percent for the overall venture sector.
• 2) The number of cleantech IPOs is lagging and their aftermarket performance is poor.
• 3) Cleantech companies tend to take longer to IPO or get acquired and require more capital to get to profitability than say, internet startups.

IPOs and long term investors

So what to do about the problem? I spoke recently with Mona Defrawi, the founder of Equidity, a startup that has built a platform to connect promising later stage startups with buy side investors. These are the investors who would typically buy IPOs, but who now are getting access to data on later stage startups so they can consider investing in growth companies a couple years pre-IPO at attractive valuations while giving up some liquidity by doing private deals.

Defrawi has some strong opinions about how the overall IPO system is broken, much of which she blames on decimalization, short term trading, Sarbanes Oxley, and the inability to have stable markets post IPOs because there aren’t enough long term investors sticking by companies through an IPO. The net result is a world where it takes close to ten years to go public versus about 5 years in the 90s, something that has a rough impact on venture capital firms that need liquidity and need to show LPs a return. Defrawi put together an infographic to detail her view of the IPO market and to promote Equidity:

Equidity launched its list of 135 GrowthSTARS in February. These are companies that Equidity wants to be the foundation of its list of companies that it can connect with long term investors pre-IPO. Scanning the list, it included a few cleantech names such as Bloom Energy, Opower, oDesk, and BrightSource. Some of the companies on the list, like Opower, I wouldn’t think would have any trouble finding capital given its continued success. Though other names like BrightSource, which specifically has had to ditch its IPO plans as it struggles with the competitive move to solar PV technology, may need capital to survive before it ever reconsiders a public offering.

But regardless of how attractive the various companies on Equidity’s list are, the real point of interest is that Equidity wants to offer up these companies to buy side investors before the IPO, particularly in a world where venture capital has grown somewhat tight.

There are other efforts to increase liquidity pre-IPO from the likes of SecondMarket and AngelList. Equidity is another such effort though it’s trying to do so at a much larger investment size and not with a focus on making a market pre-IPO but on bringing the big public investors into private deals pre-IPO.

And if Equidity is one small step toward making it easier for cleantech companies to get later stage capital and attract investors that will stick with the company post-IPO, that could aid a recovery in the upstream venture capital that’s needed to finance the next generation of cleantech startups.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• The perils of cleantech investing: KiOR and the long-term, high-risk view
• Defining success for cleantech companies
• Defining success for cleantech companies

Tesla is still a few years away from being able to sell EVs to the mass market but the company continues to plug away toward the goal. Wall Street did not like Tesla’s Q4 results, causing the share price to shed about 7 percent. The quarterly loss was worse than analysts had expected despite revenue coming in at $306 million, $7 million north of the $299 million consensus estimate.

Getting to gross margin

The earnings and revenue numbers highlight a core theme for 2013 and Tesla, which will revolve around getting to the 25 percent gross margin that Musk has promised, a milestone he reiterated on the investor conference call last Wednesday. For the fourth quarter, the gross margin was just 8 percent, owing to the high expense of ramping production and reaching economies of scale for a factory that’s geared to produce just 5,000 cars a quarter.

The small volume production of the Model S makes improving margins challenging. During Q4, Tesla flew in tires from a supplier in the Czech Republic in order to meet production goals. Air freight costs about ten times shipping freight for an automotive part, and Musk noted that the costs made him want “to punch myself in the face for that one.”

Musk told the anecdote about the Czech tires to highlight that he believes a lot of excesses like overtime and supplier costs can be dealt with over time to expand margins. When Tesla reaches certain volume levels on its orders for batteries, for example, Panasonic cuts back its pricing.

Surprises: Foreign demand and leasing

And just as Tesla has some work to meet expectations in terms of gross margin, there are also potential surprises in the company in terms of demand. The European and Asian markets open up later this year as Tesla’s head of retail George Blankenship said he’s looking forward to stores in Hong Kong and Beijing.

Additionally, the Model S cannot currently be leased, which likely is a barrier for some customers who aren’t sitting on the cash to buy the luxury car. I’d expect that to change by the end of the year, though Musk was clear that he wants compelling interest rates before rolling out a program, something I suspect could be challenging. No one really knows what a Model S will be worth in 4 to 5 years when a lease would be up. It’s such a new product for an luxury EV market about which future vehicle values are mostly a mystery.

So it gets hard for any financial institution that would underwrite a leasing program to take on risk by offering lower interest rates.  But we’ll see. I wouldn’t be shocked if Tesla is willing to offer a bit of a higher interest rate for leasing, knowing that in a recovering economy we’re going to start to see more widespread demand for luxury cars.

The Future

Finally, buried at the end of the conference call was a reference to the Model X, Tesla’s crossover SUV. Musk was reticent to give out numbers on early reservations for the all electric SUV, but he estimated that buying interest would be around 70 percent of that for the Model S. Delivering the Model X in 2014 is another incremental step for Tesla but it’s an important one and given how hard it was for the company to ramp production on the Model S, producing an SUV at good margins could be very difficult. Not to mention that significant R&D and selling expenses will be incurred to launch the Model X.

Tesla remains a production supply constrained company as it lacks the facilities to meet demand. The average wait time for a Model S is 5 months. And while Blankenship and Musk talked up potential future growth in Europe and Asia, it’s irrelevant in the short term because there are limits right now on how much Tesla can scale given that it can only output 20,000 cars a year. But all in all, these are good problems to have.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• How the Model S will change Tesla and the EV market
• The fourth quarter of 2012 in cleantech
• Cleantech and investment in 2013

And while Tesla became associated with its ultra high end sports car, the Roadster, the vision for the company has always been to bring electric vehicles to the mass market. I caught up with Motor Trend’s Editor in Chief Ed Loh to discuss the challenges ahead for the leading EV company and whether Tesla really can capture the attention of a broader automotive market.

(This interview has been edited for length and clarity).

Q).  What were the key distinguishing features that helped the Model S win Car of the Year?

A). In all of our six criteria (performance and intended function, engineering, efficiency, safety, value, advancement in design), the Model S blew us away. It’s a good looking car. It’s very sexy. It’s got a great profile, reminiscent of the Jaguar XS or the Porsche Panamera. Yet it’s an electric car, one of these new green cars, but it’s not one of these hippie-mobiles like the Prius with a weird campback profile and small wheels. In terms of design it’s gorgeous.

When you talk about engineering excellence, it’s a seven passenger vehicle. Five adults and two children can fit and it has an amazing amount of cargo space. It has a front trunk and a rear trunk. It’s a great driving car. There’s no penalty here. It’s the fastest American built sedan based on our numbers. 0-60 in under 4 seconds for the signature Performance model and it also handles really well. And on top of it, it has this fantastic electric technology. Well you look at the criteria, it’s all of these that hit it out of the park.

Q). Tesla currently has 13,000 reservations for the Model S, it’s still a very small company. Is this the beginning of Tesla becoming a more mainstream auto brand?

A). We give the Car of the Year the award. We don’t give the business case of the year award. We have some industry heavyweights that are guest judges and that was a big part of the discussion. This is a very small company and some of the numbers don’t shake out. But that’s not what it’s about. The award is about the car.

We do think this is a significant turning point. Motor Trend’s been around since 1949, 64 years of giving this award, and the award has always gone to a car powered by gasoline with an internal combustion engine. That we gave it to an electric car is a big deal in our history.

At Tesla you view this as validation from someone in the establishment. Elon [Musk] and I were chatting and one of the things we said is that we hope this is an inflection point, a turning point in the history of the automobile where new technology is being embraced. I think Tesla is definitely banking on this as something that changes people’s minds.

Q). Tesla is using a retail experience model rather than a dealership model. What is Tesla doing differently and do you think it’s going to work?

A). It’s a little early to draw any conclusions. It’s fascinating. The veterans are incredibly cynical because Tesla has positioned itself head to head against NADA (National Automobile Dealers Association). They’re challenging them because auto company owned stores or dealerships are against the law. A lot of people want to put Tesla out of business simply because they’re operating these retail outlets. Even though from Tesla’s point these are not traditional dealerships, you can’t walk in, buy a car and drive it off that day.

Obviously George’s (Blankenship) impact is clear. These are Apple stores for Tesla. They’re situated in high traffic, ritzy high end locations. I went to the opening of the one in Santa Monica and George told me, “yeah I remember ten years ago standing here on the opposite side of the street just over there” and pointed at the Apple store. They want that Apple experience… you’re dazzled by the hardware, and you get to touch it and interact with it. The key difference is you can buy an iPad. This builds the excitement.

My core concern is will enough Americans be accepting of this purchasing processing? We’re famous in the world for wanting it yesterday and buying whatever’s in stock. I spoke to a guy at Rolls Royce and he said for the American customer, the dealer has so much power because it orders all the cars. When you’re a Rolls Royce customer you can customize it to the nth degree yet nobody in America does that whereas in China and Europe the customer is happy to wait for the car and get it exactly the way they want. The millionaires in America walk in, want a white one, here’s the check, and I want the keys right now. With that kind of purchasing history mindset, will they accept a Tesla store that requires you to wait three months to get your car?

Q). Do you think now that the Model S is a bonafide competitor for the German automakers?

A). Yes and no. Until they’ve achieved volume, which is a considerable feat, the answer is likely no. It’s a rounding error for the 5 series BMW or the E class. They’re not afraid if a few people go with a Model S.

But, absolutely when you look at the specs, dimensionally, performance and in terms of cachet. It is a premium luxury sedan, it’s right in the wheel house of the Porsche Panamera. It really does span the full scope of an E class or a 5 series. It’s got better economy than a [BMW] 528 yet it performs at the level of an M5.

Q). Tesla has had some issues scaling and is behind schedule. How much of an issue do you think scaling the manufacturing process will be?

A). That truly is one of the core questions that came out of our deliberation process. Building one is a feat in itself and something they should be lauded for, but getting scale is an entirely different shooting match.

They’re not as supplier-constrained as a typical auto manufacturer. They do a lot of the building in house. Elon has put it that they bring in raw metal and plastic pellets and spit out a car. They’ve brought in a lot of what were the traditional relationships with vendors because the parts are just not available. To get up to a larger production scale, they’re already facing challenges. They are late. Any new technology is a painful process.

Q). Tesla has laid out an ambitious roadmap. It’s talking an SUV, the Model X, for 2014, for 2015 a scaled down Model S for around 30K, another Roadster and crossover SUV after that. Do you think Tesla can succeed in making EVs viable in the market?

A). The major signpost will be once the smaller battery versions of the Model S, the 40KW and 60KW models, start shipping. Obviously the Model X will be a huge step in the right direction in terms of greater volume because that’s a huge market segment.

Truly the lower priced, the mass market electric car is the biggest goal. If they can hit that and get into the Prius category, like a loaded Prius, you’ll start to see a major shift. It’s actually pretty easy to build a Bugatti Byron Veyron, a 3 million dollar no limit supercar. It’s a lot harder to build a Camry or a Prius, that gets purchased by everybody because it has all the right features at the right price. Tesla’s demonstrated they can build something at a high end, high level. If they can bring something out that can bring out the masses, that’s a big challenge.

As for the phase after that, a new supercar or a smaller SUV, those will not happen unless they can get through the Model X and the mass market electric car.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• The fourth quarter of 2012 in cleantech
• Cleantech and investment in 2013
• Cleantech third-quarter 2012

“There was a challenge around getting clean power,” says Nelson. Utah has less than 3 percent renewable energy and generates 82 percent of its power from coal, according to data from 2009 from the U.S. Energy Information Administration.

The question of how eBay would find clean power for its Utah data center — which is part of eBay’s global consolidation of its data centers into just three mega data centers in Utah, Arizona and Nevada — opened up another possibility: would it make more sense for the online marketplace to power its Utah data center by generating its own power and use the grid as backup?

The fuel cell finds a home

For the much larger Topaz data center, which is modular and built to scale to 30 megawatts, eBay has issued a 10 megawatt public RFI in the past couple months, seeking proposals to generate renewable energy for the data center. But for the adjacent smaller add on Quicksilver data center, eBay decided to have 6 megawatts of Bloom Energy’s solid oxide fuel cells installed on site, making the installation the country’s largest non-utility fuel cell installation. Bloom’s fuel cells generate power by combining natural gas and oxygen to produce carbon dioxide and water. EBay is operating a microgrid on site that will allow it to divert excess power from the fuel cells not used by the Quicksilver data center directly to the larger Topaz data center.

Nelson liked the design of fuel cells themselves because there are thousands of redundancy points built in. Each fuel cell produces 25 watts, is combined into a stack producing 40 kilowatts, and is part of a brick producing 200 kilowatts. If one or a group of fuel cells fails, the entire fuel cell brick doesn’t go down.

“Fuel cells are primary power for the computers and then the grid is used as backup. We have a higher available data center with this design than we would have with a traditional generator and UPS design. That was a big ah-ha moment for me,” said Nelson.

Aside from the redundancy of the fuel cells, Nelson was much more excited about the capital expense savings eBay’s getting from shifting his backup power source.

“We’re not putting UPS (uninterruptible power supply) or generators at Quicksilver. They’re completely removed,” says Nelson, referring to the Quicksilver data center, the smaller of the two data center projects in South Jordan, Utah. “This required us to completely change how we approach data center design. We’re using the fuel cells as the backup and the primary source.”

Nelson added that while the up front costs of fuel cells creates a scenario where the fuel cell power is more expensive than grid power, there were a number of cost avoidance returns from the fuel cells that made the economics work. Namely the elimination of UPS and generators along with the simplification of the building design that included changing the height of the building as well as the structural support needed. He noted that eBay examined the risk of an outage on the gas grid, and found it more reliable than the electrical grid. EBay also looked at large scale battery storage, which Nelson said wasn’t cost effective.

Generators and UPS exist at data centers in the event of a power outage and are used about 1 percent of the year. Generators themselves have been criticized. For example, the environmental impacts of Microsoft’s diesel generators in Santa Clara, California and Quincy, Washington were recently highlighted in a controversial New York Times article on data centers.

In terms of the broad pressure on leading webscale IT companies to move away from non-renewable sources of energy like coal power and diesel generators, fuel cells can be renewable if they use biogas. Biogas is reclaimed methane (CH4) from landfills or water treatment plants, rather than mined natural gas. It’s this reason why Microsoft has discussed the possibility of locating future data centers near water treatment plants, which wind up having to flare excess methane because it’s a 20 times more potent greenhouse gas than CO2.

There is currently no biogas produced in Utah, according to Nelson, and eBay is offsetting its natural gas use in Utah by paying a premium to enable biogas production elsewhere. Nelson said the offset will be “meter in/meter out,” meaning for whatever amount of natural gas eBay uses at its Utah data center it will enable that amount of biogas production elsewhere. EBay didn’t disclose the premium it’ll pay.

When the grid becomes backup power

Generating one’s own power is typically more expensive than grid power, which utilities generate at scale and from inexpensive coal, natural gas and nuclear energy. But there’s a growing amount of discussion from data center operators that being in control of one’s own power generation could lead to better uptime.

The June power outage in Virginia caused service problems for Amazon Web Services, impacting Instagram, Pinterest and Netflix. In places like India where the grid is inherently unreliable, companies have been generating their own power for years. Five of India’s biggest electricity consumers generate 96 percent of their own power, which is why major Indian conglomerates like Reliance Industries and Tata Motors were largely unimpacted by the massive blackout over the summer.

While getting off the grid was not eBay’s primary motivation for the fuel cell installation, Nelson did say, “If there are limitations within the grid, we’d like to not be bound to them. And we got the benefit of having the grid as backup.” Using the grid as backup creates a relatively safe redundancy, and if it enables the elimination of UPS and generators, data center operators can rethink the up front capital costs of installing costly on site power generation.

Bloom Energy

EBay is getting its fuel cells from Bloom Energy, which targets corporate customers interested in generating their own power on site. To address the hefty up front capital costs of buying fuel cells, the fuel cell pioneer now offers energy-as-a-service in which Bloom carries the upfront purchase cost of the fuel cells in exchange for a kilowatt hour power purchase agreement. We’re seeing this financing strategy in solar as well, as IPO hopeful SolarCity is actively pursuing a strategy in which it waives the costs of rooftop solar panels in exchange for a long term power purchase agreement. There’s even talk on Wall Street that bankers want to bundle the contracts and securitize them as debt products, to be known as solar backed securities though the market for fuel cells is likely too small for this type of securitization.

EBay actually opted to pay the upfront capital costs of the fuel cells combined with a 20-year guarantee from Bloom of minimal power generation, including maintenance obligations. The eBay deal is a major win for Bloom, which has dived headlong into the data center market this year, hiring Peter Gross to head up its data center group. Gross led data center design for EYP Mission Critical Facilities, which HP bought in 2007.

According to my colleague Katie Fehrenbacher’s August interview with Bloom Energy CFO Bill Kurtz, Bloom remains “halfway” to breaking even, and is in the process of raising another $150 million at a valuation of $2.7 billion. If successful, the raise would bring the lifetime capital raised for the eleven-year old company to $800 million. And it would also make Bloom number four on the top 10 U.S. venture-backed companies in terms of total equity raised, just ahead of Metro PCS and Facebook, but two slots behind Solyndra, which clocked close to a billion dollars in funding.

For the time being, Bloom will have to sell more data centers VPs on the idea that the extra expense of fuel cells has secondary benefits not evident in the initial cost of power. Like a decreased need for UPS and generators, and maybe even a data center less vulnerable to grid outages. And if the possibility of using reclaimed biogas ensures that the fuel source is renewable, all the better. For now all eyes will be on eBay, as well as Apple which is deploying a smaller fuel cell facility in North Carolina, to see how the fuel cells perform and whether they could be part of a rethinking of how to provide backup power for the modern data center.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

• Locating data centers in an energy-constrained world
• By the Numbers: What a Bloom Electron Is Worth
• Bloom Energy and Data Centers – Perfect Together?