Duncan Wingham, chief executive of the Natural Environment Research Council, also told the House of Lords Science and Technology Committee that such a decision would increase concentration as the research base “contracts around excellence”.
It would also prevent new areas of scientific strength developing and make maintaining large capital assets difficult, he said.
The peers were collecting evidence on research priorities ahead of the publication of a new government strategy for science and innovation expected in the autumn.
When asked about the possible implications of a flat-cash settlement, Luke Georghiou, vice-president for research and innovation at the University of Manchester, warned: “If it goes too far we could see a cliff edge in terms of the great results of [science].”
He said that during the current spending review, universities had managed to maintain productivity through concentration. Research-intensive universities had been able to increase funding at the expense of other institutions, he added.
“A degree of concentration allows access to economies of scale but it is not a sustainable position in the long term. Universities are having to generate surpluses elsewhere to maintain capital assets,” he explained.
Professor Wingham said: “On the resource side it will increase concentration. We will contract around excellence.
“A [research] council may be faced with significant and difficult decisions as to whether it can sustain its big capital assets,” he added.
Professor Wingham also said that research councils risk generating “desire-led science” if they continue their emphasis on cross-disciplinary projects that span all councils in the next spending period.
He explained that “desire-led science” sees researchers trying to tackle a problem that they want to solve even if science may not have an answer.
Projects may “be more effective” if they encompass just two or three different research councils, he added.
But Professor Wingham denied that research funding decisions had become too focused on economic benefit. “We could go too far, but I am not sure that is the case today,” he added.
