International Grocery Retailers' Country Environment, Resources and Local Performance. A Cross-classified Multi-level Approach

Abstract

This study aims to shed light on the often-neglected effects of host country environments on firms’ local performance. We chose this topic, because host country environment is known to affect firms’ internationalization decisions (e. g., market selection or entry mode choice) while it is rarely regarded as an antecedent of local performance after a market entry. Still, country environments are expected to have substantial performance implications by international business scholars, while we know from national research that local performance is attributed to the local environment, especially in the retail industry. We hence aim to contribute to international retailing research by conceptualizing and testing such effects across nations. Because retail firms’ resources may help to overcome environmental dependencies, such resources are conceptualized as moderators of the effects of the host country environment on local performance. We thus make a second contribution by providing novel insights on the context specificity of resources. We chose the grocery retailing industry to analyze these topics for several reasons. First, grocery retailing is an important industry (e. g., market volumes of 770 billion USD in the US, 680 in China, or 280 in Germany) which is highly sensitive to host country environments. Furthermore, in this industry different levels of the environment need to be differentiated: country-level environment specific to all firms and store format-level environment specific to store formats like hypermarkets vs. discounters. We thus consider three levels of analysis: store-formats, countries and firms. We apply cross-classified multi-level modeling, which allows us to treat the complex data structure appropriately and has rarely been used in business literature so far. A Bayesian estimation is used. We thus make a methodological contribution to literature, by showing a way to treat such data structures appropriately. The results show that purchasing power and rule of law (country level) enhance local performance, whereas local intra-format competition (format level) diminishes it. Important resources in retailing (firm level) do not affect local performance directly but the results support a context-specificity of resources, i. e., a moderation role of the environment-local performance links. In particular, country level effects are moderated by retailers’ degree of internationalization, while surprising interactions occur of the format level environment. We provide important insights for managers and applicate the novel cross-classified multi-level approach that accounts for increasing complexities in data structures in international business research.