Food Superstores, Food Deserts and Traffic Generation in the UK: A Semi-Parametric Regression Approach

This study contributes another route towards explaining and tackling ‘food desert’ effects. It features the estimation of a (semi-parametric) trip attraction model for food superstores in the UK using a composite dataset. The data comprises information from the UK Census of Population, the NOMIS (National Online Manpower Information System) archive and traffic and site-specific data from the TRICS (Trip Rate Information Computer System) databases. The results indicate that traffic to a given food superstore, ceteris paribus, increases with household car ownership, store parking provision, site size (floor space), and distance to the nearest competitor. Furthermore, increases in public transport provision are shown to be associated with increasing car trips. This latter effect is discussed in the light of planning policy for development control purposes and a role linked to the reinforcement of ‘food deserts’. The results also reveal activity-specific household economies of scope and scale. It is suggested how these may also further perpetuate unsustainable development and ‘food desert’ characteristics.