DARPA et al do not represent an analogous scenario. As for portfolio theory: I'm not questioning something that fundamental. I'm saying the government has little to no hope of deploying it efficiently. If you had a technocracy of highly intelligent people perfectly resilient to corruption, then this might be a different story. And if you know anything about VC, you know that there's a long tail of failing ones that underperform standard benchmarks (you probably do know; that's why you said "top VCs"). Also, if you want to view USG market interventionism as an aggregate portfolio, guess which way that's going to go.
I love it when people champion entities like Tesla benefiting from incentive structures when it's conveniently aligned with their personal beliefs about the superiority of the tech industry, their political agendas, and/or the idolzation of people like Musk, but forget that there's a far broader and more insidious reality here.
Reply to Below: Not libertarian. I believe strongly in government regulations where appropriate and that limiting our personal liberties in certain ways is essential to guaranteeing economic and physical security. I happen to agree with the sustainability agenda as well. I disagree with the mechanism being employed to advance it. USG should instead focus on taxing/penalizing for carbon emissions as a negative externality, which would have an ancillary benefit of making alternative energy companies overall more competitive. Asking the government to make decisions about structuring incentives, by contrast, is positively asinine (you get things like ethanol). Furthermore, I have a serious problem with the government deploying tax dollars to directly benefit corporations when there is no immediate and tangible return, regardless of the context.