Restaurant Density and Delivery Speed in Food Delivery Platforms

Abstract

With the rapid growth of food delivery platforms, understanding their operational implications for partnered restaurants is crucial. This paper compares three main externalities induced by an increase in restaurant density on a platform: customer expansion, cannibalization, and delivery pooling (i.e., drivers being in closer proximity to restaurants). Using a theoretical model, we develop testable hypotheses about the impact of these externalities on various aspects of partnered restaurants, including delivery speed (i.e., order wait time), sales, revenues, and customers' spending behaviors. Econometric analysis, based on data from a major Chinese food delivery platform, shows that as the platform's restaurant density increases, delivery speed, sales, and revenues improve for partnered restaurants. However, customers are found to opt for smaller order sizes in denser areas. The empirical evidence then indicates that delivery pooling emerges as the dominant externality compared to customer expansion and cannibalization as a platform expands regionally. Exploiting structural estimation, we further demonstrate that the driver arrival speed at a restaurant and restaurant density follow an inverted-U-shaped relationship, suggesting a diminishing benefit to delivery speed from increasing restaurant density. Lastly, using instrumental variable estimation, we find that both restaurant sales and customer spending behaviors are highly sensitive to delivery speed, underscoring the dominant role of delivery pooling. Our findings highlight the importance of optimizing restaurant density for food delivery platforms undergoing expansion and recommend a focus on maintaining high restaurant density areas to maximize delivery speed. This can foster a scenario beneficial not only for platforms but also for restaurant partners and customers.

Keywords: Food-delivery platforms, restaurant density, platform expansion, delivery efficiency