International debt and emerging market crises: Theory and empirics

The term structure of yield spreads of risky sovereign bonds reflects market perceptions of the risks of default. An inverted curve prior to default can convey important signals of the markets' consensus regarding default probabilities in the short term. I study the term structure of international bonds when default is a possibility and see how it may provide us with information about the imminent crises. I find that using the information embedded in the term structure allows us to distinguish between anticipated versus unanticipated default episodes. For episodes that are anticipated, the yield spread is sharply inverted prior to default and the default probability structure is declining in maturities.; Chapter 1 documents the various default episodes and the emergence of the bond market as an important source of external finance for emerging markets. I use the term structure of sovereign bonds to extract market-implied estimates of default probabilities for emerging markets. Using an empirical implementation of a discrete-time intensity based model, data on sovereign bond prices and the US bond yields, I derive the market implied default probabilities for six emerging countries that have defaulted on their external debt in recent times. These parameters are then used to derive the defaultable term structure. The empirical results help to document certain regularities in not only the default probabilities across countries but also in the variation of the credit spreads across maturities.; Chapter 2 presents a stochastic small open economy model with endogenous default risk and term structure, where the interest rates vary according to the probability of default. The model predicts that default incentives are higher in recessions, which is observed in the data. The paper characterizes equilibrium country interest rates, maturity profile of debt and the relation with output. A key result of the model is that agents expect a default to occur in the period prior to when the actual event can take place. This heightened default probability in the short term is reflected in an inverted term structure for the yield spreads. The implications of the model are consistent with key features of the empirical findings.