Neighbourhoods in Urban Economics: Incorporating Cognitively Perceived Urban Space in Economic Models

This paper proposes that urban economic analysis would benefit from the use of cognitively perceived neighbourhoods, which are discussed within urban studies. Georeferenced data should be aggregated by spatially bounded units that are identifiable by citizens in order to enrich one-dimensional distance as the sole variable to bring urban complexity into economic models. Multivariate analysis is applied to data from Belo Horizonte, Brazil, to formulate four indices, ranked by neighbourhoods that together represent a spatially complex, non-linear influence on urban real estate markets. The results of the indices by neighbourhood are then tested against a traditional specification in an econometrics exercise that does not include the concepts and indices put forward. The definition of neighbourhood used and the empirical results provide a thorough description of urban fabric that can fully and more accurately represent urban influence in economics while avoiding abstract distance measurement.