Property markets are frequently disaggregated into regions that commonly form the basis for property portfolio diversification strategies. However, these regions are defined geographically, rather than on the basis of functionality. Each region comprises heterogeneous local property markets. This thesis focuses on the retail sector. Specifically, temporal and spatial differences in rents are examined. Rental levels are determined in the user market, by the interaction of demand and stock. Demand is a function of rental and economic factors and, in the retail sector, local and national economic factors influence rental determination. These factors are examined. Two issues are addressed. First, given the long-standing call for property diversification strategies to move away from the regional classification and towards a grouping of functional local markets on the basis of the underlying drivers of performance, this thesis examines the development of a new classification of local retail property markets. This classification is developed using retail rental growth rates, one of the drivers of total returns. Second, the development of models of rental levels, for example, is essential to further our understanding of the drivers and structure of property markets. These models can also play a vital part in the management of property and property portfolios. Thus, this thesis seeks to develop a model of rental levels in local retail property markets. Cross-sectional analysis is used to examine a large sample of the major retail investment markets in the UK. These markets are diverse in terms of geographical location, size, position within the retail hierarchy and the socio-economic and demographic characteristics of the catchment population. This thesis concludes that, despite the diverse and heterogeneous nature of local retail markets, classifications can be developed to provide an input into the development of sophisticated diversification strategies.