Interesting article on how the UK's Research Excellence Framework evolved. An interesting comment is: "Per pound distributed, the RAE and REF are vastly cheaper than distributing the same sums via grant applications to the research councils" Very significant funding is tied to the results of the REF. By contrast, the Australian system is primarily driven by winning grants and then getting overheads based on them sent to institutions by the government a couple of years later with a smoothing process over time. PhD completions also feature strongly. So far, the ERA is only linked to a very small allocation of funding.