Hefce says that "it will be reasonable to measure relative levels of excellence on a common basis once normalisation by field has been performed". This assumes the instigation of a metrics-based RAE will not result in universities destabilising the meaning of a "field" to optimise performance.
For example, the two Proceedings journals of the Royal Society have impact factors of 1.338 and 3.612 for mathematical, physical and engineering science and biological sciences respectively. This is close to the factor of 3 difference cited as reflecting international excellence, so surely managers will make engineering and physics departments hire medics and biologists to benefit from this lower citation norm?
This will then require subjective decisions to be made by Hefce for each paper on whether it is truly a piece of physics/engineering or is in fact medicine. The lack of a single norm in a field due to varying citations in subdisciplines will mean that UK science will lose capability in a swath of important research areas as departments seek only to hire in the high- citation subdisciplines.
Chris Keylock, Leeds University.