I am in the process of writing some proposals and am encountering the fun task of generating budgets for those proposals. Rutgers, like many cash-strapped schools, imposes a hefty “overhead” charge on federal grants (the so-called indirect costs) amounting to something like more than 50% of the value of the grant. Since I’m primarily a theory guy, the largest line item on any grant I write is generally a graduate students. With stipend, tuition, fees, and benefits, a calendar-year appointment for a graduate student costs around $90k, factoring indirect costs. Given that an NSF Small award caps out at $500k, it’s quite difficult to support more than one student for a small grant. This in turn limits the scope of research one can propose — it’s all fine and well to say there are 15 journal papers’ worth of results stemming from your great ideas, but 3-4 student years is probably not enough to make that happen.
I know some schools offer a tuition break for RAs/GSRs, but I am not sure how prevalent this practice is. So I put it to the readers of the blog: what is the line-item cost to support a graduate student for one year (without travel etc.) at your institution?
An Ergodic Walk 
Graduate students are about the most expensive thing you can budget for. Post-docs are actually cheaper, since they get paid a bit more, but there’s no tuition. Our overhead isn’t quite as high as yours, but the most famous institutions have even higher indirect rates. The less “cash-strapped” a school already is, they more they get to take off the top.
The variability is incredible, a lot coming in from the tuition break or not! I’ve encountered numbers that are 56K per year to 75-80K per year. The overhead percentages can vary greatly too – 45%, 55% or even 60% in some places. While this does make a big difference, it is the tuition break or not and the extent of it, that makes a huge difference.
BTW, one more figure that I heard recently 47K per year. BTW, all of these are from high-ranked places.
Grad students at my institution (major research university that’s a state school) are expensive. On a recently submitted NSF grant proposal, a grad student year cost ~$65K (factoring in all the stuff and indirect costs). I am not sure why there would be a ~$25K difference between your institution and mine, but my guess is that cost of living in my modest midwest town is less than in NJ, and tuition might be less. My indirect cost is 51%.