[stevenbedrick]

You're right, it doesn't make sense, but I find it entirely plausible. Welcome to academia; nobody said it was supposed to make sense (especially when it comes to the financial aspects of the gig). Ask me sometime about how my university calculates overhead costs & indirect percentages, let alone how it allocates and charges for floor space- the words "byzantine" and "Machiavellian" come to mind.

Any scientist will tell you that the most important person in their department- by far- is their finance person. They're the ones that know your university's financial system inside and out, and so will be able to help you keep your money from going straight up the university's nose (as it were). A good finance person will also know how, on a per-agency-basis, to structure a grant's budget so that you get to keep as much money as possible as freely-usable as possible (E.g., there are things that the NSF will let you do with certain kinds of grant money that the NIH won't, and vice versa, and a good finance person knows what they are).