Essays on international portfolio choices and capital flows

Abstract

The goal of this thesis is to study the international portfolio choices of countries in an asymmetric world. In practice, this corresponds to the salient facts of country portfolios and the underlying structural asymmetries between developing and developed countries in a financially integrated world. In the three main chapters of the thesis, frameworks are developed to advance our understanding of the way various country asymmetries contribute to the emergence of these persistent phenomena in international capital markets. The first essay studies the question of why developing countries experience net equity inflows and bond outflows while developed countries experience net equity outflows and bond inflows, the so-called ‘two-way capital flows’. The analysis is based on an open-economy New Keynesian model of endogenous country portfolios with representative agents in each country. The model is so general that it allows one to perform an assessment of the roles of a long list of country asymmetries in determining the pattern of two-way capital flows. While steady-state net country portfolios are zero in the first essay, the second and third essays consider the situations where this is not true. The second essay presents an OLG model of an endowment economy with a country asymmetry in households’ patience. Global imbalances in net positions emerge. Gross portfolio positions are obtained as the sum of standard self-hedging and, moreover, the hedging due to external imbalances. The valuation effects of external adjustments between creditor and debtor countries are rationalized. By introducing non-tradable risks, the third essay models a production OLG economy with a country asymmetry in wealth division. Global imbalances in net positions again arise. Gross portfolio positions are composed of self-hedging, hedging of non-tradable income and hedging of external interest payments, which accounts for the reality of asymmetric asset home bias, i.e. although assets are locally biased everywhere, the pattern is more pronounced in creditor countries.