The proposed NSF budget is a vivid illustration of the agency's disfunction. Misallocation of funds and wrong priorities are much worse than the anticipated budget cut.
In Biology, we have to look at ourselves to see the cause of the dearth of curiosity-driven science. According to Bruce Alberts, Marc Kirschner, Shirley Tilghman, and Harold Varmus (2014),
“the system nowfavors those who can guarantee results rather than those with potentially pathbreaking ideas that, by definition, cannot promise success. Young investigators are discouraged from departing too far from their postdoctoral work, when they should instead be posing new questions and inventing new approaches. Seasoned investigators are inclined to stick to their tried-and-true formulas for success rather than explore new fields.”
"Young investigators are discouraged from departing too far from their postdoctoral work, when they should instead be posing new questions and inventing new approaches."
Another problem tangential to this is we let way too many people into grad school and you have a lot of young investigators with freshly minted PhDs and who actually cannot pose new questions nor invent new approaches, in their post doc or beyond. Those individuals may be able to get funded doing a lot of incremental stuff just to get a salary and call themselves a scientist, but the funding spent on their work is not facilitating major strides.
"Pathbreaking work" does not seem to be for the many. Some promise of success also seems reasonable when money is spent, so someone would have to make that determination. Difficult when path-breakers are often identified after the fact.
Is there an overproduction of academics in STEM? Just curious.
Klainerman makes some good points. A few observations follow.
1-When one is fixing a large mess at speed, one will make mistakes, which hopefully will be corrected in the near future.
2-"the reasonable intention of the Trump administration to cap indirect costs¹ for universities, often exceeding 60- 65% of a grant, to 15%, or to reduce support of activities not directly tied to research"
This is a serious problem that needs fixing. When I consider contributing to a charity, I look at what percentage of contributions go directly to support the charity's mission, as opposed to administration, publicity, ... If it's less than 80%, they get no contribution from me. If the government acts in a similar way with taxpayer funds, cheers to them!
3-Trump's huge support of US AI in all its aspects sends the most accurate signal re his view of STEM, I would say. My estimate is that once things settle (it's very early days in his administration), STEM will get its fair share, and more, of federal funds given its tremendous impacts in all areas of life, including the economy.
4-Klainerman mentions the importance of math research. Possibly the best example of this importance is number theory - the most pure (i.e. "useless", in some minds at least) of mathematics. For eons, it had no/almost no practical applications. Then computer cryptography became very important, and number theory became ESSENTIAL to creating and maintaining computer and network security via cryptographic algorithms.
To correct some fundamental misconceptions about the overhead rates:
The often-quoted "65%" creates the illusion that merely 35% of the grant is dedicated to research. In fact it is the other way around: to the "direct cost" (research funds) of the grant counted as 100%, NSF adds the overhead (say, for UC Berkeley, 57%) which goes to the university. That is, 57/157 = .36 of the grant supports the institution, and .64 supports the research.
In reality the overhead portion is even lower, because in its calculation some parts of the "direct cost" are excluded. Most notably, it is the cost of equipment in access of $5K, and in experimental sciences such equipment could be a major part of the cost.
Another excluded part is "tuition" (received by the university anyway) for the graduate students employed for research under the grant.
Finally, the rates exceeding 65% exist, but apparently only for medical schools.
Those indirect costs can still be very substantial amounts. Furthermore, misallocation or cross-subsidization (e.g. STEM grant used to benefit non-STEM departments) are not acceptable in my view.
awarded NSF post-doctoral fellowships in Math and Physics Sciences based on criteria that included advancing DEI goals. For example, the UC Berkeley Math Dept hosted a BLM Reading Group and other ideologically motivated activities explicitly funded by such an award.
The proposed NSF budget is a vivid illustration of the agency's disfunction. Misallocation of funds and wrong priorities are much worse than the anticipated budget cut.
In Biology, we have to look at ourselves to see the cause of the dearth of curiosity-driven science. According to Bruce Alberts, Marc Kirschner, Shirley Tilghman, and Harold Varmus (2014),
“the system nowfavors those who can guarantee results rather than those with potentially pathbreaking ideas that, by definition, cannot promise success. Young investigators are discouraged from departing too far from their postdoctoral work, when they should instead be posing new questions and inventing new approaches. Seasoned investigators are inclined to stick to their tried-and-true formulas for success rather than explore new fields.”
"Young investigators are discouraged from departing too far from their postdoctoral work, when they should instead be posing new questions and inventing new approaches."
Another problem tangential to this is we let way too many people into grad school and you have a lot of young investigators with freshly minted PhDs and who actually cannot pose new questions nor invent new approaches, in their post doc or beyond. Those individuals may be able to get funded doing a lot of incremental stuff just to get a salary and call themselves a scientist, but the funding spent on their work is not facilitating major strides.
"Pathbreaking work" does not seem to be for the many. Some promise of success also seems reasonable when money is spent, so someone would have to make that determination. Difficult when path-breakers are often identified after the fact.
Is there an overproduction of academics in STEM? Just curious.
I love this quote.
Klainerman makes some good points. A few observations follow.
1-When one is fixing a large mess at speed, one will make mistakes, which hopefully will be corrected in the near future.
2-"the reasonable intention of the Trump administration to cap indirect costs¹ for universities, often exceeding 60- 65% of a grant, to 15%, or to reduce support of activities not directly tied to research"
This is a serious problem that needs fixing. When I consider contributing to a charity, I look at what percentage of contributions go directly to support the charity's mission, as opposed to administration, publicity, ... If it's less than 80%, they get no contribution from me. If the government acts in a similar way with taxpayer funds, cheers to them!
3-Trump's huge support of US AI in all its aspects sends the most accurate signal re his view of STEM, I would say. My estimate is that once things settle (it's very early days in his administration), STEM will get its fair share, and more, of federal funds given its tremendous impacts in all areas of life, including the economy.
4-Klainerman mentions the importance of math research. Possibly the best example of this importance is number theory - the most pure (i.e. "useless", in some minds at least) of mathematics. For eons, it had no/almost no practical applications. Then computer cryptography became very important, and number theory became ESSENTIAL to creating and maintaining computer and network security via cryptographic algorithms.
To correct some fundamental misconceptions about the overhead rates:
The often-quoted "65%" creates the illusion that merely 35% of the grant is dedicated to research. In fact it is the other way around: to the "direct cost" (research funds) of the grant counted as 100%, NSF adds the overhead (say, for UC Berkeley, 57%) which goes to the university. That is, 57/157 = .36 of the grant supports the institution, and .64 supports the research.
In reality the overhead portion is even lower, because in its calculation some parts of the "direct cost" are excluded. Most notably, it is the cost of equipment in access of $5K, and in experimental sciences such equipment could be a major part of the cost.
Another excluded part is "tuition" (received by the university anyway) for the graduate students employed for research under the grant.
Finally, the rates exceeding 65% exist, but apparently only for medical schools.
Those indirect costs can still be very substantial amounts. Furthermore, misallocation or cross-subsidization (e.g. STEM grant used to benefit non-STEM departments) are not acceptable in my view.
Absolutely. My correcting of the misleading percentages does not invalidate your points or main messages of the article. In fact NSF directly funded DEI-like programs in STEM. Namely, MPS Ascend program https://www.nsf.gov/funding/opportunities/mps-ascend-mathematical-physical-sciences-ascending-postdoctoral/505879/nsf23-501
awarded NSF post-doctoral fellowships in Math and Physics Sciences based on criteria that included advancing DEI goals. For example, the UC Berkeley Math Dept hosted a BLM Reading Group and other ideologically motivated activities explicitly funded by such an award.
Besides all this the Simons foundation allows only 15% for indirect costs. I see no reason why NSF allows for far more than that.
Thanks.
Thi kind of thing.
https://x.com/DACDAC4DAC/status/1945160311083835712
yikes!
The termites of Crit Studies never sleep, take a day off or stop multiplying.
What will be left after they've deconstructed everything?
Fascinating read thanks.