Part of it is that research grants are not used fully as a research funding source - a typical university administration will skim about a quarter of every grant for "administrative costs." It's not called out as corruption because it's the norm, but it does have the effect of reducing the amount of tax-funded R&D dollars that actually make it to R&D
As a result, the people in charge of hiring and firing have a self-preserving interest to value grant-earners
It'll never happen, but if funding agencies like the NSF or NIH put strings on the funding like "100% must go to the PI awarded this grant" with accounting requirements, it would help remove some of the financial incentive.
It would also help lower some of the pressure to publish or perish, since a lot of that comes from the need to chase grants.
As a result, the people in charge of hiring and firing have a self-preserving interest to value grant-earners
It'll never happen, but if funding agencies like the NSF or NIH put strings on the funding like "100% must go to the PI awarded this grant" with accounting requirements, it would help remove some of the financial incentive.
It would also help lower some of the pressure to publish or perish, since a lot of that comes from the need to chase grants.