Motivation for the Research
World capital markets have experienced large-scale sovereign defaults on a number of occasions, the most recent being Argentina's default in 2002. While Argentina, which has experienced five default or restructuring episodes in the last 180 years, may be an extreme case, sovereign defaults occur with some frequency in emerging markets. Other characteristics of emerging markets are that defaults occur in equilibrium, interest rates and net exports are countercyclical, and interest rates and current accounts are positively correlated.

This paper develops a quantitative model of debt and default in a small open economy to match the above facts, with the aim of explaining the dynamics that produce these characteristics.

Research Approach
The authors develop a model of a small open economy that receives a stochastic endowment stream and trades a single good and a single asset, a one-period bond, with the rest of the world. To emphasize the distinction between the roles of transitory and permanent shocks, they present two extreme cases of their model. Model I represents the case in which the only shock is a transitory shock around a linear trend; Model II represents the case in which the trend itself is stochastic. The model is solved numerically, using the discrete state-space method.

To improve on the results, the authors augment Model II with a phenomenon observed in many default episodes-bailouts. They model bailouts as a transfer from an unmodeled third party to creditors in the case of default.

Key Findings

• The model's ability to match the facts in the data improves substantially when the productivity process is characterized by a volatile stochastic trend as opposed to transitory fluctuations around a stable trend.
• For models with purely transitory shocks, default is a rare event, as it occurs on average only once every 1,250 years. This, in turn, leads to a counterfactually stable interest-rate process. Moreover, the cyclicality of the interest rate is the opposite of the cyclicality of net exports, while in the data both are countercyclical and positively correlated in emerging markets.
• The rate of default increases by a factor of ten when the model with the stochastic trend is substituted for the model with transitory shocks. Furthermore, both the current account and interest rates are countercyclical and positively correlated.
• Allowing for bailouts of fairly modest levels compared with those observed in practice enables the model to match the extreme rates of default observed in many Latin American economies. The model also matches the countercyclicality of net exports and interest rates. However, by breaking the tight link between default and interest rates, the model fails to produce reasonable volatility in the interest rate.
• While the predictions remain short of matching the magnitudes shown in the data, the predicted signs of the correlations of income, net exports, and the interest rate are in line with empirical facts.

Implications
The reason a model with trend shocks performs better than one with transitory shocks is that in an environment with trend shocks, a given probability of default is associated with a smaller borrowing cost at the margin. This, in turn, rests on the fact that trend shocks have a greater impact on the propensity to default than do standard transitory shocks, making interest rates relatively less sensitive to the amount borrowed and relatively more sensitive to the realization of the shock.