There is "no case" for adopting a broad policy of concentrating research funding on the basis of institutions' historical standing, a new report says.
The analysis, due to be published on 25 March, uses citation data to show that when the five "golden-triangle" institutions - the universities of Oxford and Cambridge, Imperial College London, University College London and the London School of Economics - are removed from the Russell Group of large research-intensive universities, the 1994 Group of smaller research-led universities outperforms it.
The report, Research Selectivity and UK Performance, is published by the Higher Education Policy Institute and authored by the data analysis firm Evidence. It says: "There may be an argument for more selective funding to the rare peaks of internationally outstanding excellence. What seems clear, though, is that there is no case for a general and wider policy to concentrate funding based on historical characteristics."
The report praises the changes to the research funding formula for 2010-11, which will lead to more money flowing to the elite Oxford-Cambridge-London grouping, while maintaining the principle that funding should be based on merit.
It also reveals some uncomfortable truths, attributing the exceptional international impact of the UK's research to a "very small number of individuals" in an "even smaller number of institutions".
More than 20 per cent of the UK's research output goes uncited, it says, and even among the golden-triangle institutions, the figure can be as high as 15 per cent.
The majority of research done in the UK is "mediocre at best", the report concludes, although it adds that this would be "true of a profile for any other country".
"It seems that a significant amount of the research done even in golden-triangle institutions may be considered to be not very good at all, and that the money provided for such research could be better, or at least as well, spent elsewhere," it says.
The report also provides data that indicate that, for all the problems with the research assessment exercise, the UK's research performance improved substantially after it was introduced.
This improvement is strongest among Russell Group institutions, although it is a trend that has tailed off in recent years.
Bahram Bekhradnia, director of Hepi, said: "Intuitively people know it, but this shows the introduction of the RAE giving rise to a big and continuing boost in research quality...There has been a continuing improvement in the UK, whereas the US seems to have fallen back."
The main message of the report, he said, was that the UK's very highest-quality research outputs were concentrated in the golden-triangle universities.
"Beyond that, good research and modest research are both quite widely spread...The idea of concentrating research funds based on the historical characteristics of universities makes no sense," Mr Bekhradnia asserted.