Discussion on “gross capital flows Dynamics and crises” by Broner, Didier, Erce, and Schmukler

International capital flows have always been an important topic in international economics. Capital flows help a country exploit better investment opportunity and smooth consumption. Over the past few years, global current account imbalances have been at the forefront of policy debates. Large amount of research devotes to the study of net capital flows, while fewer works put their focus on gross capital flows. This paper fills the gap. This paper provides a systematic analysis over business cycle behavior of gross capital flows. It looks at 103 countries between 1970 and 2009. The countries are divided into three groups, lower-middler-income, upper-middler-income, and high-income countries. Three forms of capital flows are examined, direct investment, portfolio, and other investments. To study impact of crisis over gross capital flows, it also classifies financial crisis into different types: banking crisis, currency crisis, and domestic debt and external debt crisis. The paper has the following key findings. First, both gross capital inflows and outflows increase relative to GDP trend over period of time. Second, gross flows are larger and more volatile than net flows. Third, gross inflows and outflows are positively correlated, especially for high-income countries. Lastly, both inflows and outflows are procyclical and shrink during crises. Decline of capital flows during crises is observed in all forms, direct investment, portfolio, and other investments. Broner et al. raise a new challenge for the international literature. Standard international real business cycle model as in Backus et al. (1992) mostly has one asset and cannot address issues related to global capital flows. This paper calls for a unified theory to model gross capital flows and to account for the key findings in this paper. One candidate is the literature on portfolio choices. This literature mainly focuses on long-run composition of portfolios, the so-called home bias puzzle in asset. French and Poterba (1991) first pointed out American held roughly 94% of their equity wealth in the US stock market. Heathcote and Fabrizio (2007) give a review over the findings in this literature. Two exceptions, Hnatkovska (2010) and Tille and Wincoop (2010), study cyclicality of gross capital flows. Their models, however, generate counterfactual capital flows as pointed out by Broner et al. In this discussion, I focus on new theories that have the potential to explain co-movement of capital inflows and outflows.