Can Structural Models Explain Prices of Sovereign Bonds?

Abstract

We test the ability of an extended structural model, originally proposed by Cathcart and El-Jahel (2003), to capture the dynamics of prices for Mexican Brady bonds. In this framework, default is triggered either when a latent variable measuring financial distress falls below a specific threshold (as in structural models), or when a hazard rate causes an unexpected jump (as in reduced form models). Using market prices and a Kalman Filter methodology, we estimate the model and extract the implicit "distance-to-default" over a seven-year period. The model is slightly superior to one which assumes that distance-to-default follows a random walk. However, the hazard-rate feature of the model makes no contribution to explaining the dynamics of market prices. We find that three economic factors explain approximately 70% of the variation in the distance-to-default, namely: the level of the stock market, the exchange rate and the risk-free term structure. When the distance-to-default is approximated from these variables and substituted back into the models, the Cathcart and El-Jahel model still performs better than the naive model, not only in-sample but out-of-sample as well. The structural model is therefore supported over simpler alternatives, but only by a small margin.