[the8472]

1. diminishing returns. you don't allocate all your resources to fixing the most pressing problem right now. doubling the budget would only yield some fractional improvements. That doesn't mean that today's allocations are optimal (by which measure?), but neither would be scuttling various research programs to shift the money to humanitarian aid and agriculture. If you want to divert budget from somewhere you probably need to find the most bloated allocation that already is far into diminishing returns territory

2. long-term vs. short-term planning. we do basic research to improve something in the long term. you don't want all your resources allocated to fixing short-term problems because then you'll have a problem in the future since nobody laid the groundwork to solve the future's problems. space exploration is basic research that pays off in moderate ways today (new materials science, satellites) and hopefully in larger ways in the future, e.g. asteroid mining, planetary defense against asteroids or establishing permanent self-sufficient settlements somewhere else in the solar system.

3. you can't just blame the government for allocating resources in a way that does not seem "moral", you also have to look at the people who prefer to allocate money to such things like commercial TV as you mention yourself instead of solving more basic problems. People's utility functions and their optimization choices are complicated. There's a locality bias (self > monkeysphere > in-group > out-group) and the question how they weight and aggregate over multiple variables (geometric or arithmetic mean, medians, minimum, weights...) so as cynical as it may sound, a new smartphone or a flag on the moon is more important by some factor than feeding a million africans. People may say otherwise when asked directly, but their actual decisions when not having to make that tradeoff directly means they'll choose to use their money that way.