If UK research funding policy has sacred cows, the dual-support system must have almost as much claim as the Haldane principle to such bovine divinity.
The government received widespread support for its decision in the most recent spending review to maintain the roughly three-to-two split between the total annual research council budget - worth about ?2.6 billion - and the quality-related funding budget allocated by the funding councils, the English portion of which amounts to about ?1.6 billion.
But amid rumours that the next spending review is to be brought forward to next year, questions are being asked about whether the sacred cow should be as sacrosanct as its devotees claim.
David Price, vice-provost for research at University College London, said the popular defence of the status quo was partly motivated by a lack of understanding of how the dual-support system had given rise to such a high-quality and efficient research base in the UK.
"There is an argument that the dual-support system is magic: people don't know why it is as effective as it is, and therefore feel that to perturb it is dangerous," he explained.
Professor Price said he was partly convinced by that argument, but noted that the time and effort required to scrutinise every grant application made the research councils' distribution of funding particularly inefficient. And given the increasing struggles of the research councils to cope with demand, he believed there was a strong argument for shifting 10 to 20 per cent of their funding to the QR pot.
"This would require research councils to do less micromanagement of small grants," Professor Price said. "For example, the Arts and Humanities Research Council's Connected Communities programme is putting out calls for ?20,000 grants, with institution-level quotas. It must cost at least that to create and process the applications - it's crazy."
Professor Price would like to see the research councils focus on directive mode and larger grants, "where the cost of administration is more in proportion", leaving curiosity-driven research and strategic institutional objectives to be funded through QR.
The ability to use QR funding - which arrives as part of universities' annual block grants - to make strategic investments in their research capabilities is highly prized by most vice-chancellors. But Kieron Flanagan, lecturer in science and technology policy and management at the University of Manchester, said he was unconvinced that universities were "very good at strategy".
He was sympathetic to the idea of shifting funding away from small project grants, but he noted that universities would not be obliged to use any corresponding increase in their QR allocation to offer such grants internally. Nor would even the largest research-intensive institutions be able to match the depth of expertise available to the research councils to carry out "quality control" of prospective research programmes - "unless the universities called on external peer reviewers or advisory boards, in which case the effect would simply be to contract out the existing [research council] process to the universities".
Adam Tickell, pro vice-chancellor for research and knowledge transfer at the University of Birmingham, said most academics he knew regretted the loss of "intellectual creativity" resulting from the research councils' reductions in their spending on responsive-mode grants. But he recognised that any reduction in the research councils' share of the science budget would be "devastating" given that their existing commitments and declining real-term incomes meant most were expecting to be able to offer only very few new grants beyond 2013-14.
Such a situation would be likely to hit the sciences particularly hard as they rely more heavily on research council support. A recent Institute of Physics analysis estimated that 80 per cent of physics funding was derived from the research councils; in contrast, the AHRC estimates that it provides just 14 per cent of funding for humanities subjects.
Concerns about the fairness and administrative burden of the research excellence framework, on the basis of which QR funding is distributed, also generate scepticism about moving funding away from the research councils.
'Worst of all worlds'
For Tim Blackman, pro vice-chancellor of research and scholarship at The Open University, the contraction of the research budget has combined with the current system's bureaucracy to create "the worst of all worlds".
"The REF will involve a huge commitment of academic time to assessing outputs, impact and environment on a five-point scale when only the two highest grades are likely to be funded. Research council applications entail another huge amount of academic time reviewing applications when most will not be funded," he said.
The fact that the REF was retrospective also meant there was no guarantee that the resulting QR distribution would "support [the most] excellent research". For this reason, he supported a "substantial" shift of research funding towards the research councils, to allow the funding of "as many excellent proposals as possible, wherever they may be".
"A QR-type budget should be retained to provide a basic funding floor, but at a much lower level of funding and linked by formula to research council income in a similar way to Higher Education Innovation Funding. Future REFs could then be abolished," Professor Blackman said.
For Ian Walmsley, pro vice-chancellor for research at the University of Oxford, the REF's chief failing is its necessarily domestic focus. "Testing ideas in an international arena is key to maintaining things at a level of excellence and creativity that is sustainable in the long run," he said.
If research funding were distributed entirely on the basis of the REF, "the inward-lookingness would end up atrophying the international competitiveness of [the UK's] research activity".
Professor Walmsley also valued highly both the necessity imposed on academics by the grant application process to "sit down and ask 'why am I doing this?'" and the opportunity to have their answers rigorously reviewed by their peers.
Therefore, he believed that any change should tip the research budget in favour of the research councils. He admitted that QR was a necessary safety net for those good researchers who failed to secure research council grants given that, unlike in the US, most UK disciplines were not blessed with multiple sources of grant funding. But this paucity was a situation he would like to see addressed by the creation of new sources.
Professor Walmsley added: "A number of vice-chancellors would shoot me if I advocated the abolition of QR and putting all research funding into another set of research councils, but it would be interesting to consider whether that American model would be the right one for the UK too."