This week's announcement that the research councils will use potential economic impact as one way of deciding which projects to fund is welcome. It should lead to better decision-making, and it will allow applicants for funding to be clearer about what is expected of them.
But assessing the possible impact of a research proposal will be no simple task for research council assessors. Academics who study innovation rarely agree on the economic effects of past research, let alone work that has not happened yet.
And despite research council reassurances, the idea of assessing the possible economic impact of research is bound to be troubling to the vast majority of academics, from historians to botanists, whose work never produces a patent and who never launch a spin-off company.
As attempts to measure teaching and research quality in British universities have shown, professors who scent money are adept at finding the right things to say about what they are doing. So it is inevitable that claims about the economic impact of research programmes will become a routine part of funding applications. In some cases, perhaps a research project on a specific new material, it may even be possible for some relatively definite numbers to be attached.
But society faces many challenges. It is important to avoid a narrow definition of the economic benefits that research can bring. In the era of climate change, almost any research into Earth systems might have a massive impact. With ageing populations, anyone researching the brain can say they are fighting degenerative disease. In the UK, one of the world's top tourist destinations, museums and galleries need research to stay ahead of the competition, just as manufacturing industry does. Even the astronomers will claim they are saving us from mass extinction by asteroid strike.
This would not matter if research councils had plenty of money. But they fund fewer than a third of the applications they receive. Using economic impact as an extra measure of research quality risks favouring short-term projects over longer-term ones, especially if the proposal is written to tap into some current government concern.
The next stage is for the research councils to be more specific about what they mean by economic impact. It is normally seen in terms of engineering and science. But modern societies are not so simple. A successful economy needs good schools, so we need education research. We need cleverer and better-informed governments, which suggests that economics and other social sciences ought to be supported. But these broad considerations are not much help if we are trying to choose between specific projects whose impact will take years to become apparent. The new system will also need flexibility. Religious fundamentalism and obesity are two issues that have become prominent in recent years and are now regarded as central to the future of Britain. Both would have been thought peripheral a few years ago.
The research councils also risk misjudging the mood of the Government by putting too much stress on the direct economic value of the work they support. Gordon Brown is known to be keen on research because it is something the UK is good at. Its excellence - rather than its economic spin-off - gives the Prime Minister bragging rights on the international stage. Perhaps more importantly, blue-skies research of no apparent commercial use captures the imagination of the young and encourages them into difficult subjects such as science and engineering. The main contribution universities make to the economy is to supply it with graduates. So research that engages the public by its inherent interest might be the most valuable of all.