Explaining the latest insanity

Reporters generally don't know much, if anything, about the subjects they cover, and pundits generally know nothing at all about the subjects of their bloviations. To some extent, this can't be helped. They are required to be generalists, and they can't be experts on everything. There are exceptions who cover specific fields. There are some good science writers, people who understand military strategy, Krugman of course who actually understand economic reality unlike most economists. Obviously that means most pundits are actually useless.

Anyway, I'm not seeing any adequate explanation of the sudden announcement by president Musk that he's going to restrict NIH indirect costs to 15%. I have to assume that more than 90% of people have little or no idea what this means, and the people who write about it for the New York Times and CNN definitely don't (Fox News reporters know nothing at all about anything, they just make shit up), so let me tell you. 

Research institutions -- mostly universities, including medical schools, schools of public health, and biology departments -- depend on NIH and other federal sources for most of their funding.  Smaller federal sources include the Agency for Healthcare Research and Quality, National Science Foundation, Advanced Research Projects Agency, CDC, and functional departments that fund some research related to their activities including the Department of Defense, the Agriculture Department, NASA, and others, but NIH is the most important. There is some private foundation funding for biomedical and public health research, but it amounts to a quite small percentage. (Right now, 100% of my own funding is actually private, but I'm semi-retired and I'm not bringing in the kind of bucks that make my bosses like me.)

Federal grants consist of two parts, called direct and indirect costs. Direct costs are pretty much what it sounds like -- the incremental expenses necessary to conduct the research. This includes salaries for the investigators, research assistants, data analysts, and other personnel who do the tasks to get the answers. It may also include equipment, supplies, use of computer resources, compensation for research subjects, travel to scientific conferences -- whatever is spent for the specific project.

Indirect costs are how research institutions pay for their ongoing infrastructure. That means buildings, utilities, administration; physical and technical infrastructure that is shared among researchers such as the Institutional Review Board, tech support, and grant administration (which is quite complicated because federal proposals and contracts are extraordinarily elaborate); and yes, the research institution's share of the university's overhead including the arguably inflated salaries of the president and other potentates, the faculty lounge, and the homecoming parade. However the latter is pretty small. (We negotiate it with the university.)

Each institution negotiates its indirect cost rate with the federal government. The same rate applies to every grant, which may or may not make sense but it makes things easier. The typical rate for a research university is above 50%. Last I checked ours was something like 62%. That means if I get a grant for $250,000, the university actually gets $405,000. Maybe that sounds unreasonable, but it's the result of a process in which the university spelled out, line by line, what it actually costs them to house an investigator -- offices, support staff, utilities, expenses for recruitment and hiring, information technology, grant and contract administration, all the stuff I mentioned above; and federal bureaucrats looked at the number, asked questions, raised objections, and ultimately set a rate which was somewhat less than the university asked for.

Does that mean there's no possibility it might be higher than it needs to be? Certainly not. Bureaucracies tend to metastasize, and get involved where they don't really need to be. I could do with a bit less of it. But simply declaring, across the board, that we're going to slash the rate by 75% is, well, insane. The research enterprise cannot possibly survive that. It will put biomedical and public health research in the U.S. out of business because it will be economically impossible to conduct it. Nobody will apply for federal research funding because it will cost them more to try to do it than the government will pay. That's what this means -- Musk wants to shut down extramural research supported by the federal government. That will be more "efficient." 

Just so we're clear.