---

Universities: Research Assessment Exercise

My Lords, I thank the noble Lord, Lord Williams of Elvel, for initiating this debate. It is timely for all the reasons that have been given. I also congratulate my colleague, the noble Lord, Lord Rees of Ludlow, on an excellent maiden speech. He probes the farthest corners of the universe in his research and doubtless some of that light will shine in this place for years to come. We look forward to that. I recall the story of a series of Civil Service interviews, at the end of which the successful candidate was summoned and told that he had got the job—it was a very senior position. Just as he was leaving, he said, ““Well, is there any advice you can give me?”” The advice was: ““You are evil, but you are necessary””. That is the view that many of us took of the RAE when it was first set in place in the late 1980s. It has probably grown increasingly evil in the eyes of most. The question today is whether it is still necessary. Let us consider the situation in the late 1980s. I joined what was then called the Committee of Vice-Chancellors and Principals sometime in 1985. It had 55 members, which represented the total number of institutions at university level which were funded to do research. There are now three times that number of institutions in membership of the equivalent body, all of which might reasonably claim that they want to be funded for research. That poses the first question: affordability. Can we afford the system that we had, whereby you expected a certain volume of cash for research because you were a university of a certain size? It was seen very clearly by those who were responsible for the system that that was unlikely to continue. At that time, about 14 per cent of the age cohort went into higher education; the aspiration is now 50 per cent. That is a good aspiration, but it raises questions about the affordability of a system which allocates money for research partly on a per capita basis, as it used to do. I said ““per capita basis””. In fact, in the pre-RAE days, research money was given in a way that was not transparent; it was given in a way that was based on history—how much you had received previously; it was probably based on informed prejudice. The system’s lack of transparency probably suited the vice-chancellors because the money came in a large black sack and could be spent in the institution according to the predilections of the vice-chancellors and/or their governing bodies. That system could not continue, especially with the cost of research going up. The costs of what we call ““big science””, engineering and medical research have been rising dramatically because of their nature. That is the background to the debate. We have already heard recounted today the various ways in which the system has begun to crack, twist and turn. People do play games, and all the faults which have been attributed to it are real. However, certain difficulties are worth highlighting. The system offers returns which diminish in relation to the amount effort you put into your application for funds. The amount that moves incrementally from one research assessment exercise to another has gone down. Therefore, if you hope to raise dramatically the flow of cash coming into your institution, it is almost, if not actually, impossible. In the same way, large sums of money do not move down the system in amounts which are sufficient to create significant new centres. The amount of money moving around has diminished despite the fact the costs of operating the system have gone up. While I endorse the points that have been made by others, I mention two other difficulties which have not been mentioned so far. The humanities face a very real problem—I shall come back to that briefly. A significant problem also faces single-subject or single-field institutions. I give the example of the Institute of Education in the University of London, which is highly regarded throughout the world. In the first RAE exercise, that institution scored the highest possible score: five. It could do no better. In the next exercise, it scored five once again, but the fund reduced for the obvious reason that the pot for education had not increased sufficiently to cover the rise in quality of research and RAE applications that came into the system. Therefore, there was no way in which it could enhance its grant or even maintain the level at which it was operating. That is a particular problem for single-field institutions, of which there are quite a number. For larger institutions, in which it was my happy position to work, we could move money around between departments, faculties and subjects; there was still a degree of flexibility. But for single institutions, the money comes per capita for teaching and at that stage there is an element for research that can go down without one’s grade diminishing. I declare an interest. I am on the governing body of the Courtauld Institute of Art. I fear that it might find itself in a similar position, because however well one does, if one is a small institution with a single subject, that sets parameters for the outcome that one can hope for. We have been challenged, and rightly so, but it is not enough to show what the problems are. What are the ways in which we might go forward? I have my own few suggestions to add to those that I hope are taken into account. First, I stress that the evaluation and targeting of resource is still essential. It is necessary to have a mechanism for doing that. Secondly, judgments are required across the whole system and across all faculty areas. Metrics numbers are important, and I have no doubt that they will play an increasingly important part in the sciences and in engineering. For the humanities they are not equally effective but they are not irrelevant. The fact that someone reads one’s papers is not irrelevant to their quality. It is not the sole criterion, but it is important. Metrics have their place. Judgment should be taken into account, but on the other hand should not be deduced simply from a single set of measures or numbers, because I have no doubt that they will be flawed. Thirdly, the noble Lord, Lord Giddens, has already pointed out that a huge volume of information has been built up. We know a great deal about the research capacity of our system of universities. That has been accumulated over nearly 20 years. My suggestion is that we should take that into account in making future allocations for research. That is not shooting in the dark. It is known which institutions have a long and continuing capacity to produce high quality research. I should like to suggest that a future research assessment exercise should move away from the single up to the larger grouping: faculties, institutes and universities. It is known which universities, institutes and faculties are constantly producing high quality research—there is a track record. Yes, they can go wrong—I will return to that point in a moment—but the amount of effort taken to come to a decision on such matters compared with the amount of money that moves in the system is disproportionate. I suggest that grants for research are made to institutions over whatever length, whether it is five, seven or 10 years, but that they are subject to review at institutional and faculty level as appropriate. This is a system that the MRC and the Wellcome Trust operate extremely well. If after five years one is not producing the goods then there will be a change in one’s financial research circumstances, but there is not constant looking and shoulder leaning. Finally, if the funds are allocated in that way, it allows the possibility of institutions making judgments and backing winners that have not been picked up by the standard peer review process. The system is necessary, but has to change.