Do surges in international capital flows influence the likelihood of banking crises? Cross-country evidence on capital flows bonanzas and bonanza-boom-bust cycles

This paper asks if bonanzas (i.e. surges) in net capital flows are associated with a higher likelihood of banking crises and whether this association is necessarily through a lending boom mechanism. Using a new database covering over one hundred countries during 1973-2008, the paper shows that previous-year bonanzas in net capital flows are associated with systemic banking crises, even in the absence of a lending boom. Given a baseline bonanza, the odds of a crisis the following year are up to three times higher. The more extreme is the windfall of capital relative to trend (i.e. an intense bonanza), the larger is this risk and a crisis becomes seven times more likely. The correlation of mild bonanzas with crises is found to be necessarily associated with a lending boom; this is not the case for intense bonanzas, suggesting the existence of a different mechanism when the windfall of capital is too large. For developing countries it is found that intense bonanzas are associated with even higher odds of future crises. When decomposing flows in FDI, portfolio-equity and debt, it is found that bonanzas in both debt and portfolio-equity flows are correlated with future crises; however, portfolio flows are the ones associated with the largest increase in the likelihood of a crisis.