As a former academic researcher now managing external research grants, I have seen the shameful and growing dependence of research universities on indirect costs coming in though research grants. These supposed costs should cover the fixed expenses the institutions have in providing the researchers with support services, from office, to secretaries, to labs. These costs are fixed for each institution, ranging from 40 to over 50 cents per dollar spent on the actual research project. In practice, these funds go to all manners of other expenses, some noble, e.g. unrgraduate education, some obscenely wasteful, such as a bloated layer of vice-presidents charged with pushing researchers to bng in more research grants.
These abuses are bad for the federal government, which is overcharged; bad for state universities, as they prompt state legislators to cut funds for education, since they see all these funds as a way to avoid supprting their institution of higher education; bad for researchers, as they are evaluated on the basis of how much indirect cost money they bring in, rather than better metrics; and bad for the public, as fewer useful research projects are funded for the money spent.
The suggestion is to eliminate fixed institution-wide indirect costs. Every grant should come with its own indirect costs, detailing what support, if any, the host institution will provide: an office for the PI, computer facilities, what labs and what equipment. This info is already collected, but it bears no relevance to the indirect costs. In my experience, some departments/univerities provide researchers with a lot of support, which justifies much, if not all of the indirect costs; others provide just a desk and a kick in the rear if you don't bring in money; yet all charge rich indirect costs.
This is ludicrous and encourages all manners of abuses and bad habits, and wastes taxpayers money.