Search

This dissertation concerns cross-country consumption risk sharing in a long-run perspective. Financial integration, empirically measured by cross-country holdings of assets and liabilities, has increased dramatically in the past two decades. But what can explain the lack of cross-country risk sharing documented in the literature? Chapters 2 and 3 of this dissertation address this question. In Chapter 2, we set up a model to illustrate the mechanical difference between a bond economy and an insurance economy. We show that a bond economy can intertemporally smooth consumption in face of transitory output shocks, but not for permanent output shocks; an insurance economy is essential for risk sharing on permanent shocks. We therefore show that when both transitory and permanent output shocks exist, transitory shocks only create "noise" if the focus of interest is on identifying risk sharing in the long run. In Chapter 3, we specify an empirical nonstationary panel regression model to test long-run consumption risk sharing across a sample of OECD and emerging market countries. This is in contrast to tests in the literature which are mainly about risks at business cycle frequency. We argue that these existing tests neglected the permanent elements of risks that are of interest and that their model specifications were not rich enough to accommodate heterogeneous short-run dynamics. Since our methodology focuses on identifying cointegrating relationships while allowing for arbitrary short-run dynamics, we can obtain a consistent estimate of long-run risk sharing while disregarding any short-run nuisance factors. Our results show that, for the period of 1950-2008, the level of long-run risk sharing in OECD countries is similar to that in emerging market countries. However, during the financial integration episode of the past two decades, long-run risk sharing in OECD countries increased more than in emerging market countries. Furthermore, we investigate the relationship betw