[mattkrause]

MIT’s current overhead rate is 50.5%, but that’s pretty standard.

Here’s a scatterplot showing lots of research institutions’ rates. When this was published, MIT’s rate was slightly higher (54%). https://www.nature.com/news/indirect-costs-keeping-the-light... showing actual and calculated rates.

This also applies to federal research grants and is meant to cover costs associated with actually hosting the research (rent, utilities, support staff). Foundations can (and often do) negotiate lower rates. I’m not sure how donations are handled, but I don’t think the same F&A rates apply.

[wegs]

The article quotes 56%, as the negotiated rate on NiH projects, not 54%, but the rate varies by funder and by project. The 2/3 number is from an actual project I had insight into as the base overhead rate.

What's a little bit hidden there is that MIT dips into these funds multiple times, in pretty complex ways. For example, a sponsor might pay overhead and graduate student tuition (which just flows into MIT's general coffers). Or graduate student tuition might be waived, and the sponsor pays just overhead. Or a donor might pay overhead when the money comes in, and again on specific purchases. Or capital expenditures might waive overhead. Etc. The level of complexity is high, while the level of transparency is low.

I'm not claiming any of this is unique to MIT by any means. It's just where I have the most visibility.

What's cute is that MIT claims to lose money on everything. In the article you cited: "'We lose money on every piece of research that we do,' says Maria Zuber, vice-president for research at the Massachusetts Institute of Technology (MIT)." You'll find similar statements about educating undergraduates and tuition. And just about everything else. I've worked through the numbers at some point, and MIT only loses money with clever accounts; it's good PR to say MIT subsidize everything it does, but it's often not a reality.