He's not arguing that stars shouldn't exist; he's arguing that funding decisions should be less influenced by "star power" and there are policy tweaks that could help achieve that.
It's worth noting that at the NIH (etc), you don't really know who "the allocators" are. Program managers make the final funding decisions, but they rely heavily on peer review and you don't know who in the study section will be reviewing your proposal. Changing this so they don't know the proposer's identities either might help.
He's not arguing that stars shouldn't exist; he's arguing that funding decisions should be less influenced by "star power" and there are policy tweaks that could help achieve that.
It's worth noting that at the NIH (etc), you don't really know who "the allocators" are. Program managers make the final funding decisions, but they rely heavily on peer review and you don't know who in the study section will be reviewing your proposal. Changing this so they don't know the proposer's identities either might help.