Agent-Based Model of an Urban Retail Market

The paper presents an agent-based model of an urban retail market that explains the stylized facts about the location of shops in Brno/CZ. More specifically, it explains why some types of shops are more dominant in the center, some in the inner city, and some in the periphery of the city. In the model, consumers spend their incomes in different types of shops located randomly in a square city. The types of products sold in the shops differ in the transportation costs and in the proportion of the income spent on them. The consumers buy each type of product in the shops with the lowest price plus the per-unit transportation cost. The price of each good is determined by the profit margin assigned randomly to each shop. In each period, new shops of all types enter the market and locate randomly in the city, and some of the shops in loss exit the market. This way the urban retail market evolves toward an optimal structure; i.e. toward the optimal prices, sizes, and locations of shops. The model is calibrated using the data from the Brno retail market in 2009, and for a realistic setting of the model, it generates patterns consistent with the stylized facts.