A Discrete Choice Model for Retailer Selection in Emerging Markets

Abstract

In the context of increasing competition between chained retailers and family-owned retailers, it is key to understand the customer's format choice. Using a logistics regression (i.e., binary logit) model, we explain customers' preference to buy in supermarkets or in small-scale, mom-and-pop stores like nanostores. We collect a representative sample of over 110 surveys from customers in the district of Surco, Lima, Perú, which is a representative area of the features of Lima's residents. We asked customers to focus on analyzing their preference between two retail formats: modern channel (i.e., big-box retailers, supermarkets, and hypermarkets) and traditional channel (i.e., mom-and-pop stores, nanostores). Our surveys included factors pertaining retail format attributes as well as factors related to the purchasing process. The results showed that time available for purchase and a comparatively better perceived service at a mom-and-pop store (i.e., nanostore) are significant factors that explain a higher probability of selecting these retailers, while a better store's ambience benefits more supermarkets. The overall discrete choice model is able to explain 65% of the variance using pseudo R-squared of the actual format choice decisions.

Citation

Avila, M.A., Larco, J.A., Antonini, C., Ortíz, M.B. and Mejía Argueta, C. (2020), "A Discrete Choice Model for Retailer Selection in Emerging Markets", Yoshizaki, H.T.Y., Mejía Argueta, C. and Mattos, M.G. (Ed.) Supply Chain Management and Logistics in Emerging Markets, Emerald Publishing Limited, Leeds, pp. 299-318. https://doi.org/10.1108/978-1-83909-331-920201014