Does spatial proximity influence commuting trip length?: an approach based on evidence from Flanders and Brussels

Little consensus exists about the ideal geographical scale of the compact city model with regard to the travel patterns of its residents and users. In Flanders, the average trip distance (12.5 km) has become too large to be possibly associated with the size of the existing compact cities. However, this does not necessarily mean that the alleged relationship between the characteristics of the compact city and travel behavior is not valid. In this paper, an attempt is made to assess the impact of different measures of spatial proximity, in casu between housing and the job market, on commuting distances in a quantitative way. This is done for the region of Flanders and Brussels (the northern part of Belgium), taking into account disruptions by the modifiable areal unit problem and spatial autocorrelation.We conclude that relationships between characteristics of spatial proximity and average commuting distances in Flanders and Brussels also appear to exist at the regional scale, emphasizing the importance of a high residential density, an equal jobs-housing balance and a good accessibility of the labor market, both within and outside areas that correspond to the compact city model.Little consensus exists about the ideal geographical scale of the compact city model with regard to the travel patterns of its residents and users. In Flanders, the average trip distance (12.5 km) has become too large to be possibly associated with the size of the existing compact cities. However, this does not necessarily mean that the alleged relationship between the characteristics of the compact city and travel behavior is not valid. In this paper, an attempt is made to assess the impact of different measures of spatial proximity, in casu between housing and the job market, on commuting distances in a quantitative way. This is done for the region of Flanders and Brussels (the northern part of Belgium), taking into account disruptions by the modifiable areal unit problem and spatial autocorrelation.We conclude that relationships between characteristics of spatial proximity and average commuting distances in Flanders and Brussels also appear to exist at the regional scale, emphasizing the importance of a high residential density, an equal jobs-housing balance and a good accessibility of the labor market, both within and outside areas that correspond to the compact city model.C