Should retail stores locate close to a rival?

I simulate firms competing in an urban landscape that is segregated by income level. Some firms are located close to wealthy neighborhoods and far from competitors, but other firms do not enjoy these advantages. The probability that a household selects a certain firm to patronize depends on that firm’s price and its distance from the household. The firms have almost no information about the urban market they inhabit, but they can experiment with prices and learn what actions lead to higher profits. Firms that experience persistent losses go out of business. When transportation costs are low, households shop aggressively and the result that emerges resembles perfect competition. When transportation costs are high, households shop less aggressively, which gives firms greater market power. The results here more closely resemble monopolistic competition, with more firms surviving, higher prices and higher profits. However, this collusion is implicit, as there is no mechanism for the firms to communicate or explicitly collude with each other. I validate the model by conducting a Probit analysis of the factors that determine firm survival. I also conduct an econometric analysis of the factors that determine a firm’s price, quantity sold, and profit. Firms locating close to rival have a higher probability of going out of business. If these firms do survive, on average their profits are no different than the profits of firms that are not located close to a rival.