I have a feeling that the expected duration of games as far as in-game time was very different than what it's turning out to be given the tactics that players are discovering. If you imagine that everything was intended to be done with dispatches, and that taking e.g. 40 months to research something is a reasonable move, then some of the small numbers make more sense (mining rates are another example of this I think). But I'm getting the feeling that games tend to be much shorter - maybe 10-15 in-game years?
It'd probably be useful for rebalancing things to actually survey the player base and see how long games actually last, to figure out e.g. the net RCI effect a given building will have over the course of play.
Another thought is, for buildings like the University that give +RCI/month, what if instead the numbers were bigger but they saturated - something like +1 Economy/month so long as Economy is less than 20, otherwise +0.1/month.