Home Bias in European Countries within a Bayesian Framework

Abstract

The home bias is defined as the tendency of the investors to invest a larger proportion of their wealth in domestic equities than what would be optimal based on the meanvariance principle. There are several explanations for this observed home bias, e.g., barriers to foreign investments and information asymmetry. From a Bayesian viewpoint the level of the investors' prior mistrust in a certain asset-pricing model may explain the home bias, despite the fact that statistical tests fail in rejecting the model. The purpose is to analyze how fragile the investors' prior confidence in ICAPM (International Capital Asset Pricing Model) must be to cause home bias in European equity markets. We use a Bayesian approach to estimate the predictive distribution of the asset returns for each European country under different prior scenarios. The investors' optimal portfolio weights are constructed from the moments of this predictive distribution. The result shows that there is a strong home bias in most countries, which cannot be explained by any degree of disbelief in the ICAPM. The losses due to the holdings of inefficient portfolios by pension funds are assessed via certainty equivalent calculations. Italian pension funds suffer more than the funds of the other countries from the home bias while UK and the Netherlands experience very small losses.

JEL Classification: G11, G12