Given the many and varied uses to which journal rankings are put. interest in ranking journal ‘quality’ is likely to persist. Unfortunately, existing methods of constructing such rankings all have inherent limitations. This paper proposes a new (complementary) approach, based on submissions to RAE 2001. which is not restricted to a pre-defined journal set and, importantly, is based on quality choice decisions driven by economic incentives. For three metrics, submissions to RAE 2001 are compared with the available set of publications to provide evidence on the perception of journal quality, a fourth metric is based on the overall RAE grades, and an overall ranking is produced.