The high-profile bankruptcy of solar manufacturer Solyndra – the recipient of a $535 million federal loan guarantee from the Department of Energy – is sending shockwaves through the clean energy community, rightly so. Few other companies have been so closely identified with the Administration’s current clean energy push, or been the subject of as much controversy. The failure certainly won’t make the politics of clean energy support any easier, and the timing could hardly be worse.But the real problem isn’t necessarily Solyndra’s failure in itself. Indeed, counterintuitive though it may seem, we should actually expect some advanced technology companies receiving public support to fail. What do I mean by this? Remember a fundamental truth about innovation: when you’re trying new things, the likelihood of failure is high, and the newer and harder the things you’re trying, the more likely it is that you’re going to fail. Just ask Steve Jobs. And failure rates in the venture capital world can be quite high, after all. Unfortunately this logic doesn’t lend itself well to the politics of public investment. In general, government is most helpful when it focuses on the kind of risky innovation with big social returns that the market can’t deliver purely on its own. The tricky part is that if public innovation programs are doing what they should, and investing in truly cutting-edge technologies and firms, there should be some failures. This is because companies that receive public support should, by definition, be doing things nobody has actually tried before commercially, but nevertheless have clear upside. If innovative ventures receiving public support never failed, you’d know something might be wrong: the bets might not have been risky enough. This is why innovation programs need to distribute their bets across a wide range of emerging technologies and firms, such that the successful ones will make up for the failures. But that doesn’t mean simply propping up any company with a new idea; that risk has to come with clear upside in pushing the technological envelope, to get clean energy competitive with fossil fuels. Indeed, continuous innovation and steady performance improvements should be a fundamental condition of any public support for clean energy. But then the political system still needs to be willing to accept the failures that come with advanced technologies new to the market, which is asking a lot. When government-funded ventures do fail, opponents of public support for innovation will always be quick to make hay. The Loan Guarantee Program shouldn’t be judged solely by Solyndra, but many will do exactly that. Indeed, many innovative programs have come under fire over the years for not being risky enough, and political realities contribute to their risk aversion. There are real and numerous funding gaps for innovation in the energy market. The real challenge for effective policy isn’t to avoid failure, but to effectively distinguish between the good bets that can drive innovation but might fail, and the bad bets that should have been avoided from the beginning. And this is not a determination that should be made after-the-fact. And here is where the Solyndra situation gets complicated. The reasons for the company’s failure are complex. No doubt, China’s aggressive investments in solar competitiveness – as well as their discriminatory trade practices – helped to create a tough, incredibly competitive market environment for Solyndra. But others have argued that the company was a bad bet from the beginning, with real problems in its technology and its business model, and simply wasn’t ever going to succeed, China aside (though, of course, Solyndra’s many high-powered investors would have disagreed). Accusations have also flown that the DOE loan program wasn’t nearly rigorous enough in assessing loan recipients, technologies, and market trends, and was instead simply trying to push stimulus dollars out the door, to give the Administration the ability to tout green jobs. If this is indeed the case, as the House Energy and Commerce committee has argued, then the Department of Energy has some explaining to do. DOE needs to prioritize new technologies that have pathways to competitiveness or can otherwise contribute meaningfully to moving the cleantech ball down the field, and that requires some hard-headed realism, which may have been in short supply during the rush to stimulate the economy. It also requires managing expectations, which suggests the Administration should be careful about making green jobs claims. A big question is whether Solyndra’s technology was innovative enough to merit public largesse. Maybe, maybe not; I will leave it to others to figure out those details for sure. But the bottom line is really this: failure is not the enemy. Avoidable failure is the enemy. Fear of failure is no reason to roll back vital public support for innovation and competitiveness, and using failure to argue against support for innovation gets it exactly backwards.