Maybe, though in STEM fields, anyways, the Department pays for the first year of tuition (essentially the University paying itself), and than an advisor pays ou

Maybe, though in STEM fields, anyways, the Department pays for the first year of tuition (essentially the University paying itself), and than an advisor pays out of his grant money (which is usually from the Fed. gov't, and so essentially the Feds paying themselves). So in a perfect world, Universities would just say grad-student tuition is 0$ for the first year, since they don't net any $ from it anyways, and than the Federal gov't would just up grants by whatever is needed to cover the extra tax bill, which would be less efficient but ultimately a wash as far as the money is concerned. In reality, I think this will happen to some extent, but not enough to totally reverse the change, and there will be a sizable drop in grad-students pop. in the US. Given extra grant money, Profs are likely to use some of it to cover their grad-students tax bills, but at the same time, those students will now be more expensive relative to fancy microscopes or whatever, and so fewer grad-students will be hired.