We probably want even more R&D than we have now. Private companies frequently spend much less than the optimal level because it's a tossup on whether it will amount to anything, government can fund a wide range of R&D and society picks up the benefits of it even if half the projects don't turn out.
