The effectiveness of macroprudential policies and capital controls against volatile capital inflows

Focus

Countries around the world use macroprudential policies and capital restrictions as a policy response to volatile capital inflows. This paper asks: are these policies effective in reducing or changing the composition of gross capital inflows, the probability of capital inflow surges, and the probability of currency and banking crises?

Contribution

The authors assess the effectiveness of policies with a model that takes into account the likelihood that countries will implement a new macroprudential policy or capital control. In particular, they look at macroprudential policies that target foreign currency borrowing and inflows, and compare their effects with those of other macroprudential policies and controls in the following year. They do this for a large sample of countries over nearly two decades, allowing for high-level evidence on effectiveness.

Findings

The paper finds that capital inflows into countries are lower when those countries use macroprudential policies that target foreign currency, but not after changes in capital controls. Macroprudential policies can also lower the probability of so-called surges in capital inflows, and of banking crises in the following three years. There are no significant effects on the probability of currency crises. There are no significant effects of capital controls. Overall, the results indicate that macroprudential policies - especially those that target foreign currency mismatches - may be more effective at responding to volatile capital inflows than capital controls that discriminate on the basis of residency.

Abstract

This paper compares the effectiveness of macroprudential policies (MaPs) and capital controls (CCs) in influencing the volume and composition of capital inflows, and the probability of banking and currency crises. We distinguish between foreign exchange (FX)-based MaPs, which may be similar to some types of CCs, and non-FX-based MaPs. Using a panel of 83 countries over the period 2000-17, and a propensity score matching model to control for selection bias, we find that capital inflow volumes are lower where FX-based MaPs have been activated. The imposition of CCs does not have a significant effect on the volume or composition of capital inflows. Further, we find that the activation of MaPs is associated with a lower probability of banking crises and surges in capital inflows in the following three years.

JEL classification: F38, G01, G28

Keywords: capital account openness, capital flows, capital controls, macroprudential policy, banking crises, currency crises