Dealing with dynamic endogeneity in international business research

Abstract

Dynamic endogeneity occurs when the current values of a study’s independent variables are affected by the past values of the dependent variables, which can lead to biased estimates. Our analysis of 80 empirical papers published in the Journal of International Business Studies uncovers cases of inappropriate treatment of dynamic endogeneity. Our simulations reveal factors that lead to bias in a fixed effects estimator and highlight the advantages and disadvantages of the system generalized method of moments estimator. We demonstrate our points with an empirical study of the impact of the international experience of managers and board members on firm internationalization. We show how using a fixed effects estimator rather than a generalized method of moments estimator can lead to differences in regression results. We also show the proper use of a generalized method of moments estimator.