Neighbourhood socioeconomic change is a complex phenomenon which is driven by multiple macro- and micro-level processes. Most theoretical and empirical work has focused on the role of urban-level processes, such as filtering, life-cycle, and social dynamics. For individual neighbourhoods, these processes generate flows of different socioeconomic groups, which consequently leads to an exchange of relative positions in the metropolitan hierarchy ('exchange' effect) where some neighbourhoods move up and others move down. Neighbourhoods are also affected by structural processes that operate beyond the urban level. They can generate upward or downward shifts of absolute income across a whole array of neighbourhoods ('growth/decline' effect), or change the inequality among neighbourhoods, where the top and bottom of the neighbourhood hierarchy move away from each other ('inequality' effect). A common practice in neighbourhood change studies is to represent neighbourhood status as relative to the respective metropolitan area; this neutralizes the 'growth/decline' effect and ignores an important source of change and divergence between neighbourhoods in different regions. Some specific relative measures of change do capture the inequality effect but confound the 'exchange' and 'inequality' effects. This paper introduces a methodological approach that decomposes total neighbourhood socioeconomic change, measured in absolute terms, into components of 'exchange', 'growth/decline' and 'inequality'. It applies a decomposition method presented by Van Kerm (2004), developed for understanding income mobility of individuals. The approach (1) acknowledges the role of structural processes in neighbourhood change, and (2) makes a distinction between different processes that generate neighbourhood change which is essential for comparative research.