It transfers money from grant funds into the University general fund, where it can be used for other purposes, like paying for less lucrative departments and administrative salaries. More or less, this plus the ph.d. student’s stipend is the cost of the student, and both can be charged to a grant that the student is working on.

This is baked into grant funding and has been for at least 40-50 years. No one is doing anything underhanded here; NIH is fully aware of the situation. Some grant organizations, including NSF, are much stingier with their grant overhead (ie: don’t allow it).

Whether or not the overhead is fair is a matter of opinion, I guess. I don’t think it’s even close to the most broken aspect of grant funding, though.

It’s more expensive for the prof to hire the student.

At MIT it is easily 100k per year to fund a PhD student. But phd student gets much less than half of that.