RDP 2007-11: Global Imbalances and the Global Saving Glut – A Panel Data Assessment 8. Conclusions

In this paper we have attempted to explain movements in current account balances using panel data. The model appears to capture the broad patterns in the current account balances of east Asia, China, the major oil exporters, and Argentina and Brazil over most of the sample. The model does less well from 2003 onwards and does not match the trends in the current account balances of Japan or the US.

Overall, our results provide some support for the global saving glut hypothesis. Demographic shifts and changes in fiscal balances, while statistically significant, cannot fully explain recent developments in current accounts. Expectations of future growth, as suggested by the intertemporal model of the current account, appear to influence the destination of capital flows. The ability to provide sound financial instruments, as emphasised by Caballero et al (2007), may be important, as the quality of institutions was found to have a large effect. In order to fully evaluate the interaction between supply and demand of financial assets, which Caballero et al emphasise, a more structural model is probably required. Additionally, some of the determinants of international capital flows – such as the perceived quality of financial markets, institutions, or investor sentiment – are inherently difficult to measure. Finally, financial crises do appear to increase net capital outflows from crisis regions for a number of years, although the estimated magnitude of these effects is insufficient to fully explain the pattern of recent international capital flows.