The surge of climate-tech startups seeking to reinvent clean energy and transform huge industrial markets is fueling optimism about our prospects for addressing climate change. Tens of billions are pouring into these venture-backed companies in just about every field you can imagine, from green steel to nuclear fusion.
As I explain in “Climate tech is back—and this time, it can’t afford to fail,” investments led by venture capitalists could play a critical role in developing novel sources of clean energy and greener industrial processes. Speaking to numerous VCs, people at startups, and those academics who study innovation in so-called deep tech, I became convinced we’re in the early stages of a carbon-free economy.
But the optimism comes with a warning. As a journalist who wrote extensively about cleantech 1.0, which began around 2006 and collapsed by 2013 as countless solar, battery, and biofuel firms failed, I have a sense of wariness. All of it feels a bit too familiar: the exuberance of the VCs, the hundred of millions going to risky demonstration plants testing unproven technologies, and the potential political backlash over government support of aggressive climate policies. Writing about the current climate-tech boom means keeping in mind that most previous venture-backed startups in cleantech have failed miserably.
Today’s investors and entrepreneurs hope this time is different. As I discovered in speaking with them, there are plenty of reasons they might be right; there is far more money available, and far more demand for cleaner products from consumers and industrial customers. Yet many of the challenges seen in the first boom still exist and provide ample reason to worry about the success of today’s climate-tech startups.
Here are some of the key lessons from cleantech 1.0. To learn more, you can read my full report here.
Lesson #1: Demand matters. This is basic to any market but is oft ignored in climate tech: someone needs to want to buy your product. Despite the public and scientific concerns over climate change, it’s a tough sell to get people and companies to pay extra for, say, green concrete or clean electricity.
A recent study by David Popp at Syracuse University and his colleague Matthias van den Heuvel suggests that weak demand, more than the costs and risks associated with scaling up startups, was what doomed the first cleantech wave.
Many of the products in cleantech are commodities; price often matters above all else, and green products, especially when they are first introduced, are typically too expensive to compete. The argument helps to explain the great exception to the cleantech 1.0 bust: Tesla Motors. “Tesla’s been able to differentiate their product: the brand itself has value,” says Popp. But, he adds, “it’s hard to imagine that there’s going to be a trendy [green] hydrogen brand.”
The findings suggest that government policies are probably most effective when they help to create demand for, say, green hydrogen or cement rather than directly funding startups as they struggle toward commercialization.
Lesson #2: Hubris hurts. One of the most obvious problems in cleantech 1.0 was the extreme hubris of many of its advocates. Leading cheerleaders and money men (yes, nearly all were men) had made their fortunes on computers, software, and the web and sought to apply the same strategies to cleantech.
“Rule number one: do not have people invest in a category who do not know about the category,” says Matthew Nordan, general partner at Azolla Ventures. “Cleantech 1.0 investors were largely folks from tech and biotech, desperately trying to come up to speed on industrial categories they knew little about.”
These days many venture capitalists profess to be chastened by the experience of cleantech 1.0 and deeply ingrained in the industries they hope to disrupt. But there are still some high-profile investors parachuting in from making fortunes in Big Tech who are convinced they have the solution to the world’s biggest problem.
I asked Josh Lerner, a professor at Harvard Business School who studies how venture capital works, why such investors haven’t learned from the past. The pessimistic view, he says, “is that these guys are just megalomaniac characters who want to save the world and view themselves as heroes, and they’re just fools plunging again even though they had their head handed to them before.” A more optimistic view, he says, is that they might be able “to take some of the knowledge and innovations that happened in the software arena and put them to work here.”
Lesson #3: Molecules are different from bits. Yes, of course, we know writing code is easier and cheaper than building a steel plant. But just how much riskier and unpredictable it is to scale up molecule-based businesses was an unpleasant surprise to many during cleantech 1.0. Poor yields or the synthesis of unwanted by-products—problems that might have seemed like small hiccups in the lab—can be show stoppers as the process is scaled up and must compete against existing technologies.
Finding out whether a process is commercially competitive typically means building a demonstration plant, often costing $100 million or more. Many startups during cleantech 1.0 got tripped up when processes that worked fine in the lab didn’t work nearly as well in larger facilities. You just don’t know if an industrial process will work until you build it.
The hope these days is that far more computation power and the use of artificial intelligence will allow startups to simulate how processes will work before actually building anything. Running a new way to make green hydrogen in silico to see what goes wrong is certainly far cheaper and safer than building a $100 million demonstration plant.
Lesson #4: The real takeaway from Solyndra. The failure of the company, which received a $535 million loan guarantee from the US government to manufacture a novel type of solar panel, is the one that everyone remembers from cleantech 1.0. And it’s often offered as strong evidence of what goes wrong when governments try to pick winners. But the lingering lesson from the failure of Solyndra is quite different.
First—whether you’re in government or a venture capitalist—don’t invest in technology that makes little manufacturing sense and has dubious market demand. Solyndra’s product was a highly complex cylinder-shaped solar panel that required custom and unproven equipment to build.
See lessons #1, #2, and #3. I wrote this in 2011: “What Solyndra lacked, though, was market savvy and manufacturing flexibility. Although the company had quickly traversed what Silicon Valley’s entrepreneurs like to call ‘the valley of death’—the risky financial period between receiving initial venture funding and beginning to earn revenues—it badly faltered in turning its operations into a viable, long-term business. If there is a prevailing lesson from the Solyndra debacle, it has to do with the danger of trying to do too much too quickly—and doing it alone.”
Solyndra would likely have failed anyway, but had the company gone slower, a lot of people, including both US taxpayers and the VCs who ponied up hundreds of millions, would have lost a lot less money.
Lesson #5: Politics can change everything. The 2022 Inflation Reduction Act, which helped fuel the recent wave of cleantech investments, passed Congress without a single Republican vote. Simply put, electing a Republican president in 2024 could mean an end to the aggressive federal climate policies.
And there is an ongoing backlash in many other industrial countries. Recently in the UK, the prime minister proposed weakening the country’s climate policies. Even Germany is showing signs of backing away from political support and funding for cleantech.
In his recent paper, Syracuse’s Popp and his coauthor traced the woes of cleantech 1.0 back to a largely forgotten Senate election in early 2010. After the death of the liberal Democrat Ted Kennedy, Massachusetts voters elected the Republican Scott Brown, dooming a comprehensive climate bill being debated in Congress. Without the possibility of carbon pricing, many venture investors lost interest in clean-energy startups.
By the end of the year, a newly elected Republican majority in the US House of Representatives had doomed additional large federal investments in clean energy.
Politics do matter. And they can change overnight.
Lesson #6: Survival is all about the economics. The early days of cleantech 1.0 were filled with enthusiasm and good intentions. People saw climate change as an existential crisis, and technology, led by visionary entrepreneurs and venture capitalists, was going to solve it. The vibes these days are in many ways similar; in fact, people are even more intense and committed. The brilliance of many new climate technologies is evident, and we desperately need them.
But none of that will ensure success. Venture-backed startups will need to survive on the basis of economics and financial advantages, not good intentions.
The simple fact is that we have too few examples of prosperous climate-tech startups with radical new technology. It’s all still a grand experiment. Cleantech 1.0 taught us what can go wrong. We’re still learning how to get it right.