LAUSR

Abstract While the debate around monocentric versus polycentric urban structures persists, a growing number of traditionally monocentric cities are pursuing a transformation agenda towards the polycentric structure. The decisions of locating and prioritising the additional centres is often based on an accessibility score, leaving out other essential factors such as the spatial variations of labour density and agglomeration externalities. This paper conducts an empirical investigation on the suitability of accessibility measures in identifying, evaluating and prioritising metropolitan activity centres (ACs), and proposes an effective density-based measure for the purpose. The proposed effective job/labour density measure integrates both a proximity dimension and a scale/size dimension, and is therefore more comprehensive than the accessibility measure. The two measures (accessibility and effective density) are applied to a case study of strategic-level ACs in the metropolitan region of Perth, Western Australia, and ACs are ranked based on their separate accessibility and effective density scores. Generally, the ACs performed poorly under both measures, with many non-AC zones having much better accessibility and effective job/labour density scores (than those designated as ACs). Nevertheless, the results show that the two measures produce markedly different ranking/prioritisation outcomes, suggesting that studies or policies that merely rely on the accessibility measure to make AC prioritisation (or exclusion) decisions could be misleading. By virtue of being more comprehensive with an integration of proximity and scale dimensions, and the ability to estimate economic agglomeration, the proposed effective density measure is, argued to be better suited and more reliable than a mere accessibility score.