The problem with sinking a lot of money into science is that it only pays dividends if solutions to the big problems actually exist. They may not. It may in fact not be possible to do better-than-breakeven fusion on a small (relative to a star) scale. The only way to find out for sure is the hard (and expensive) way.

By way of contrast, when it comes to military technology you're not competing with nature, only with other humans, and so a "solution" (if you'll excuse the irony) is almost certain to be found by one side or the other.