The research assessment exercise review panel wants to see radical change. The THES asks three experts to do some out-of-the-box thinking

In many ways, the research assessment exercise is a brilliant system. It is the last free lunch on public funds. There is no other mechanism whereby you are so generously rewarded on the basis of past performance rather than on what you are going to do with the money over the next five years. There is no clawback; failure, while embarrassing, is penalty free, writes Peter Knight.

Contrary to mythology, the RAE is rather popular with vice-chancellors. It is a gladiatorial contest, in which winners and losers are picked externally. Vice-chancellors can stand back from the rough and tumble of unpleasant decisions and wash their hands of departments that have failed to make the research grade.

Other advantages of the RAE include the fact that we all know the rules; there are no surprises in the way the system works. This allows the usual university policy of "thoughtful timidity" to come to the fore. We "cling to nurse for fear of something worse". The RAE also allows everyone to play, and for the post-1992 universities it is welcome as it allows newcomers to compete in the research game. So what is the problem?

ÖйúAƬ

There is a scene in one of the Indiana Jones films when Short-Round asks Indie why he is looking for ancient treasures. Indie's answer is: "Fortune and glory, fortune and glory." That is the problem with the RAE:it gives both fortune to the university and glory to the successful departments. There is every possible incentive in seeking to excel. Success means recognition in one's discipline and movement up the league tables as excellence is always interpreted as excellence in research.

The RAE is just too good at achieving the objective of driving up the standard of research. That is why the Treasury was successfully ambushed by the results of the 2001 RAE. It was not grade drift, we all simply got better. And when our collective improved performance was translated into money, it was no surprise to anyone, except the ÖйúAƬ Funding Council for England, that there was not enough to go round.

ÖйúAƬ

If the problem is with the mechanism rather than the results, the changes that need to be introduced are straightforward. We need to dampen down the system. There are lots of possibilities. The balance between the RAE allocation from the funding councils and the project funding by the research councils could be changed in favour of funding by project. It is strange that the research councils are so short of cash that many excellent projects go unfunded while RAE money continues unabated.

Perhaps some of the funding that travels via the RAE could be diverted to the Department of Health for direct allocation to research relevant to the National Health Service. Medical research should not be carried out in isolation of clinical needs. Some funds could be removed from the RAE and directed to the appallingly titled Third Leg funding stream that supports activities directly relevant to industry and commerce.

Other options include using some other measure, such as external income, as a surrogate for the measurement of research. The logic being that if someone else is giving money to a university, then it is not unreasonable for the government to add its share to meet the overheads that all projects incur.

Whatever change is introduced must achieve two objectives. First, it must take the heat out of the RAE. It needs to change the system from an object of high passion and debate to an instrument of stultifying dullness. Passion never serves universities well. Second, and here is the tricky bit, it has to give exactly the same answer as the present RAE in terms of which universities get the money. Any new system that has the effect of removing research money from Oxbridge et al is a nonstarter. It is not even worth thinking about the row that would result. Equally, any system that achieved a greater concentration of research funding would be deeply unpopular and would risk leading to the exclusion of the originality and occasional good fortune that leads to the development of new ideas and concepts. Competition is good for the academic soul.

ÖйúAƬ

So we must change the system but not change the result. An interesting challenge.

Peter Knight is vice-chancellor, University of Central England.