Comment

Abstract

Previous articleNext article FreeCommentSebnem Kalemli‐OzcanSebnem Kalemli‐OzcanUniversity of Houston and NBER Search for more articles by this author PDFPDF PLUSFull Text Add to favoritesDownload CitationTrack CitationsPermissionsReprints Share onFacebookTwitterLinked InRedditEmailQR Code SectionsMoreReinhart and Reinhart provide a systematic study of the episodes of influx of capital, namely a “capital flow bonanza.” The authors develop an algorithm, following the work of Kaminsky and Reinhart (1999), to date the incidence of bonanzas. This algorithm allows them not only to detect the smooth deterioration of the current account but also to analyze the macroeconomic developments surrounding the bonanzas. The study suggests a strong link between flows, global interest rates, and commodity prices. Using data from 181 countries from 1980 to 2007 and a core sample of 66 countries from 1960 to 2007, the authors show that (a) the path of the current account around bonanzas is V-shaped, (b) bonanza periods are associated with a higher incidence of banking and currency crises in developing countries only, (c) bonanzas precede sovereign default episodes, and (d) fiscal policy is procylical around bonanzas.I think that this paper is an extremely valuable study for anyone who is interested in capital flows and sovereign debt. My main comments will be about the data issues, robustness, and the generalization of the results. Most of the analysis is done for 66 countries, during 1960–2007: 58 are middle‐ (emerging) and high‐ (industrialized) income and eight are low‐income (Africa) countries. Are there really data from 1960–2007 for all these countries? It would be nice to indicate which countries have how many years of data in appendix table A1. If a certain set of countries have more years of data than others, then these sets might be biasing the results. More important, most of the low‐income and some of the middle‐income countries are heavily indebted poor countries, which have received a lot of aid that shows up as capital inflows. Côte d’Ivoire, Bolivia, Honduras, and Mauritius are some examples. There are also some countries with debt forgiveness that will show up as capital outflows. I think that there needs to be a robustness exercise in which these countries are excluded. At the moment it is not clear how the authors deal with these issues of aid and debt forgiveness.We might also worry about pooling emerging and industrialized countries for the empirical analysis since this type of pooling might mask important differences. Why keep oil exporters and importers together, for example? Or why force the same coefficient on eastern European countries that are low savers running a current account deficit and Asian countries that are high savers running a current account surplus? Interestingly enough, both of these sets of countries have similar average growth rates in the last decade. I think that the paper would also benefit from clarifying the definition of the bonanza further. The starting point of the paper is that the old definition did not detect smooth deterioration of the current account. However, it is not very clear how the new definition is improving on this. The reader is referred to appendix figure A1, where the labels “capital flows” and the “constructed capital flows” are not explained. The figure shows a stark difference between the “current account,” “capital flows,” and the “constructed capital flows” during 1970–74 and also to some extent the latter part of the sample. And it is not clear why this is the case.Finally, the results should be interpreted with caution since they represent conditional correlations instead of causal effects. Causality runs both ways here. The authors are indeed careful about not interpreting their regression results as casual. However, the difference between unconditional and conditional (on bonanza) probabilities of default can also be due to many other “third” factors. For example, for the United States, mortgage‐related securities are a significant part of the foreign purchases, and hence lower mortgage interest rates may be a result rather than a cause. Also in the case of procyclicality, which is based on the correlation between output and government spending, the main question is, what drives both?To sum up, I think the authors undertake a very rich and systematic empirical study that brings many issues together. There are many interesting results in the paper that highlight the importance of future research on the underlying drivers and causes of capital flows and sovereign defaults.ReferenceKaminsky, Graciela, and Carmen Reinhart. 1999. “The Twin Crises: The Causes of Banking and Balance of Payments Problems.” American Economic Review 89 (June): 473–500.First citation in articleGoogle Scholar Previous articleNext article DetailsFiguresReferencesCited by Volume 5, Number 12009 Article DOIhttps://doi.org/10.1086/595997 Views: 94 © 2009 by the National Bureau of Economic Research. All rights reserved.PDF download Crossref reports no articles citing this article.