US external debt sustainability revisited: Bayesian analysis of extended Markov switching unit root test

The sustainability of US external debt, which has been an issue of global concern, is analyzed using a Markov switching (MS) unit root test applied to the flow of debt, i.e., the current account. The first to apply the MS unit root test to the issue of US external debt in order to examine local stationarity and global stationarity were Raybaudi et al. (2004). This paper introduces an extended MS unit root test where the transition probability is time-varying rather than fixed, as is usually the case, and the change of probability is explained by the real exchange rate, which theory suggests has a close relationship with the external balance. The extended MS unit root test calculated by the Markov Chain Monte Carlo (MCMC) method provides us with new insights on the issue of US external debt in recent years, suggesting that even though the debt/current account–GDP ratio remains relatively high, the probability of stationarity (sustainability) is unexpectedly high when recent US dollar depreciation is taken into account.