I don't have my offer letter (and in any case it is a decade old), but the tuition part was definitely higher than $16k. It is possible that the sticker price i

I don't have my offer letter (and in any case it is a decade old), but the tuition part was definitely higher than $16k. It is possible that the sticker price isn't very useful, and they're taking whatever they can from outside sources and waiving the rest. There are other costs too: here's an interesting study that tries to estimate non-tuition, non-stipend costs https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4353078/#B6 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4353078/#B6 These, since they're mostly faculty time, are also partially paid for by funders. In any case, my point is that grad student labor is both important surprisingly expensive and it's not at all clear to me that we have the "right" amount of it. Specifically, I think it might be better, from a research productivity and a "system health" persepective, to have somewhat fewer grad student slots and a few more staff scientist positions: Maybe 70/10 instead of 100/~0.