The people collectively provide the initial funding when stuff is still too high risk for private investors but they rarely sees any of the upside when it hits

The people collectively provide the initial funding when stuff is still too high risk for private investors but they rarely sees any of the upside when it hits explosive growth except in a hand wavy "high tide raises all boats" or "cheaper widgets means more buying power" kind of way. In extreme cases it means we fund technologies that generate wealth that is immediately extracted by foreign entities but even in the best of cases, it creates a moral hazard that is best exemplified by the military industrial complex. It's a resource allocation efficiency issue, through and through. The government is always going to pick "winners and losers" in high value industries like defense, resource extraction, and aerospace but all of those investments have very rough feedback loops that are disconnected from the decision making process. If government investments/grants came at least with profit/revenue sharing arrangements, there would be a positive feedback loop and an opportunity for the system to evolve and improve even if the net income change after administration overhead is zero or even slightly negative. Obviously this isn't tenable for basic scientific research but our industries do very little of that anyway, even with government funding available.