Equity home bias : a global perspective from the shrunk frontier

Equity home bias research explicates the necessity for correct characterisation of benchmark domestic and foreign equity investment weights which serve as optimal weights inputs to home bias estimation equations. This paper introduces Bayesian shrinkage based improvements, to the traditional mean-variance optimisation framework for benchmark country investment weights and home bias estimation, documented in existing home bias literature. Home bias is investigated using Morgan Stanley Capital Index returns data for 39 countries in the period 2000 – 2009. We find reasonable segmentation in the global distribution of country level equity home bias. By applying the system GMM panel analysis procedure to account for potential endogeneity in determinants, we find that country idiosyncratic risk, country financial structure and country institutional quality constitute the three most important determinants of global equity home bias over the last decade.