Equity home bias : a global perspective

Abstract

Despite ongoing financial globalisation and integration, equity home bias seemingly symptomatic of investment inefficiency, continues to persist as a puzzle within the global equity investments framework. Extant research on the subject of home bias explores the phenomenon from many different angles like - barriers to international investments, hedging benefits offered by cross-border investments, information asymmetry and the behavioural aspect of "non-rational" investor behaviour - amongst others. A concern that underlies these various angles of research finds explication in Baele et al. (2007). How are the benchmark domestic and foreign equity investment weights required for the estimation of equity home bias best characterised for a reliable and meaningful explanation of the home bias puzzle? My thesis attempts to examine the issue of benchmark optimal weights characterisation by applying two new methodologies in equity home bias estimation, namely the Bayes-Stein shrinkage approach and the Black-Litterman model. The shrinkage approach reduces estimation uncertainty of the traditional mean-variance optimisation framework resulting in more acceptable equity home bias estimates. The Black-Litterman approach attempts two additional improvements - first, it provides an intuitive prior, the CAPM equilibrium market portfolio, as a starting point for estimation of asset returns, which provides an intuitive connection back to the market. Second, the Black-Litterman model provides a clear way to specify investors' views and to blend the investors' views with prior information. Research studies examining the determinants of home bias do not attempt to address a crucial contradiction underlying the "home bias puzzle", i.e. equity home bias persisting within a framework of ongoing global integration, by examining for a specific relationship between financial integration and equity home bias. Moreover there appears to be an absence of modelling studies that account for country specific heterogeneity and potential endogeneity in determinants. My thesis attempts to fill this gap in the literature by investigating, amongst other determinants, the specific relationship between financial integration and equity home bias. Within a panel framework for determinant analysis, the Difference GMM approach controlling for potential endogeneity and country specific fixed effects, and the System GMM approach controlling for unobserved heterogeneity and endogeneity in determinants, are applied. The results suggest a non-linear relationship between equity home bias and financial integration. Country economic exposure, country idiosyncratic risk, the global financial crisis and institutional quality are found to be the other important determinants of equity home bias over the last decade.

Date of Award	2016
Original language	English