The authors study the response of gross capital flows in emerging market economies (EMEs) to different global financial shocks. Their focus is primarily on the potentially stabilizing role of domestic investors, and they hypothesize that it offsets the response of foreign investors to adverse global shocks. The authors find evidence supporting this hypothesis.

The authors’ results have three useful implications for practitioners. First, they find that domestic investors play a stabilizing role when global risk-aversion shocks occur by making sizable asset repatriations that largely offset the retrenchment of foreign investors. When the global shock comes from monetary policy rather than from risk aversion, the retrenchment of foreign investors is not offset to nearly the same degree by repatriation from local investors. This result provides additional insight into the differences in behavior between foreign and domestic investors, which is important in determining the appropriate policy response when such shocks are expected.

Second, the authors find that net capital flows—that is, the difference between capital flows by foreign investors and capital flows by domestic investors—do not appear to differ according to the degree of financial integration. The dynamics of gross capital flows look substantially different, however, when comparing EMEs with high and low degrees of financial integration. Specifically, highly integrated economies are more sensitive to external financial shocks than less integrated economies. This result provides critical insights into financial market development in EMEs.

Third, the authors find that global risk aversion shocks are followed by foreign investor retrenchment across EMEs but that the decline in gross outflows is significantly larger in economies with larger reserve buffers. Thus, to the extent that this result makes domestic investors more willing to repatriate capital invested abroad during global shocks, it implies another channel through which higher reserves help increase the resilience to adverse external shocks.

The authors use data from 38 EMEs from the first quarter of 1990 to the second quarter of 2014. They also obtain data on 13 financial and real variables from the International Monetary Fund’s Balance of Payments statistics and from World Economic Outlook, Haver Analytics, and Bloomberg. These variables are then used in a panel vector autoregressive (PVAR) model, with which the authors quantify the dynamic impact of global financial shocks on both net and gross capital flows to EMEs. Two versions of the PVAR model are estimated: one focusing on net capital flows and the other focusing on gross capital flows. In order to test the potentially different impact of various external financial shocks, each model is tested using three variants: a benchmark US shocks model that tests the impact of US risk aversion and short-term interest rate shocks on capital flows, an augmented US shocks model that allows for shocks to long-term US interest rates, and a model of global shocks that redefines variables to account for reserve currencies in areas other than the US dollar.

Because the PVAR model assumes the effects are homogenous across countries, the authors conduct three additional tests to gain some insight into how heterogeneity across countries could potentially affect the results. They study differences between countries across the degree of financial integration, foreign asset and reserve holdings, and types of flow.

Many researchers in this area focus on net capital flows, which makes the authors’ emphasis on gross capital flows interesting. The sizable asset repatriations by domestic investors that offset the retrenchment by foreign investors suggest a transfer of asset holdings from foreign investors to domestic investors during global shocks, which could, in turn, help explain the volatility asset prices experience during crises.