Critics of the research assessment exercise who insisted that its results could easily be replicated through metrics must be eating their words as they pore over this week's working group report. After months of modelling, who would have predicted that most of the big winners would be new universities, or that the institutions of the Golden Triangle would find themselves at opposite ends of the scale? Only when the previous RAE results are factored in does Cambridge University escape a hefty cut, while even this would not save Imperial College London, let alone Manchester or Sheffield universities. Was this really what the Chancellor had in mind when he set the metrics hare running in his Budget? Gordon Brown, after all, was the man who poured millions into the Cambridge partnership with the Massachusetts Institute of Technology. In one way perhaps it is: the recognition of applied research (undervalued in previous RAEs) chimes with Treasury policy. For universities such as Cranfield and Sunderland to be rewarded for successful interaction with industry makes sense if the RAE is dismissed as a wasteful exercise in ossifying higher education's research effort. But for every Sunderland there is a Portsmouth University, where a growing research culture would be set back under any variant of the new system.
There would be little point (other than financial) in abolishing the RAE if metrics did no more than replicate previous results. But the modelling published this week has such a scattergun look that it is hard to imagine any of the options being adopted. By publishing them, the working group has made it harder to achieve consensus on a more sophisticated and logical version. Every university will lobby for the model that offers the best returns, and any reworking that benefits both Oxford and Cambridge, for example, will be regarded as a fix.
Nevertheless, there will have to be a reworking - probably with the incorporation of bibliometric measures and other factors beyond straightforward research income. Higher Education Minister Bill Rammell's view that research selectivity has gone far enough is both welcome and surprising, given the philosophy behind the Next Steps report, but most current models actually reverse recent moves to concentrate funding. Since the 2008 RAE is to go ahead, there is no need to rush a system through.
Indeed, the idea that metrics should be "phased in" from 2009 already risks making a nonsense of the forthcoming assessments. Far better to take the time to develop a methodology that works in principle and in practice.