UK universities are publicly funded to carry out teaching and research. Since the mid-1980s, the bulk of the research stream of institutional grants has been allocated on the basis of periodic research assessment exercises, the most recent of which was completed in 2001. The results of RAE2001 will influence institutional grants from 2002-3 onwards. This paper explains the RAE system, discusses its advantages and drawbacks, outlines a framework within which it can be analysed, and examines some of the available evidence about the impact of the RAE. The paper then concludes that the RAE system as presently operated has outlived its usefulness, and that it should be replaced by an allocation method based on the volume of research grants and contracts attracted to an institution. A short postscript updates the paper to take into account the White Paper on higher education that was published in the UK in January 2003.