Does The Difference In Valuation Between Domestic And Foreign Investors Help Explain Their Distinct Holdings Of Domestic Stocks Formerly Do Different Interpretations Of The Same Information Help Explain The Home Bias PDF Download

The basic International CAPM model with homogenous investors across the world predicts that a representative investor should always put a fraction of his wealth in equities and the optimal choice is the world portfolio. However, a large number of investors do not hold stocks at all. In addition, the aggregate holdings of domestic stocks are much higher than the share in the world portfolio. Even when investors go across borders, they prefer nearby foreign countries which have high correlations in returns with the home country. The fact that a large fraction of the population in all countries ignore the positive equity premium by not holding any equity is recognized as limited participation puzzle (Mankiw and Zeldes [1991]). The gap between actual holdings of home equities and the share of home equities in the world portfolio is documented as equity home bias (French and Poterba [1991]). The preference of nearby foreign countries which provides little diversification benefits over other foreign countries is referred as the correlation puzzle. In chapter 2, we build a general theoretical framework where these puzzling phenomena are jointly explained by wealth inequality and information inequality. The model incorporates wealth heterogeneity, entry cost and information acquisition into a classical multi-country noisy rational expectations model. It generates endogenously limited participation and underdiversified equity portfolios. The model makes various predictions that are supported by empirical evidence. It also links underdiversification and limited participation to fundamental wealth heterogeneity and information heterogeneity. This mechanism behind the scenes of underdiversification and limited participation suggests the importance of individual level heterogeneity in shaping the country level portfolio holdings. It suggests that various underdiversification phenomena share some common roots in wealth and information heterogeneity. Chapter 3 applies the multi-country model in a symmetric two-country context to study the prominent equity home bias puzzle. It is the first in the literature to show that a majority of equity home bias is attributable to the extensive margin, which we define as the proportion of equity holders who do not participate in the foreign equity market. This type of bias calls for a different type of explanation than the usual intensive margin bias, which refers to the home biased portfolio conditional on positive holdings of both home and foreign equities. This research provides an entry cost based explanation to the extensive margin bias. It helps explain the nature of overall bias by also showing the portfolio heterogeneity among participants. We incorporate wealth heterogeneity, entry cost, and information acquisition into a rational noisy expectations model to generate both limited participation and portfolio heterogeneity among participants. The model can replicate the share of foreign equities in the US equity portfolio in both margins. It makes predictions consistent with the following stylized facts: first, the participation rate is lower in the foreign market than in the home market; second, participation increases in wealth; third, the share of wealth invested in equity increases in wealth; fourth, information acquisition increases in wealth. Finally it also uncovers a new surprising fact: households with foreign equity holdings reduce the degree of international diversification as wealth increases. The past several decades have witnessed a fall in equity home bias as well as a rise in wealth inequality across countries. This research links these two phenomena. Chapter 4 gains new insight into the equity home bias puzzle by introducing wealth inequality into the discussion of cross border equity holdings. The two-country rational noisy expectations model, which features wealth heterogeneity, entry costs and endogenous information acquisition, predicts that wealth distribution plays an important role in determining foreign equity holdings. Foreign equity holders belong to the wealthy group. Hence, a country that has a higher level of wealth holds more foreign equity. Moreover, a country where the top wealth group owns a larger share of total wealth holds more foreign equities and exhibits a lower degree of equity home bias because the lower wealth inequality amplifies barrier of foreign market participation. Our empirical results support the theoretical predictions. We confirm that measures of wealth inequality in the host country are significant explanatory variables in determining foreign equity holdings: The higher wealth level is positively associated with foreign equity holdings. The measures for wealth inequality have also a significant negative impact on a country's foreign asset holdings.