Fear of disruption: a model of Markov-switching regimes for the Brazilian country risk conditional volatility

In the literature, little role is attributed to the country risk conditional volatility in the determination of the macroeconomic equilibrium in a developing small open economy (DSOE). This paper posits the prime hypothesis that, in the presence of multiple equilibria and self-fulfilling prophecies, one of the reasons why investors prefer to speculate in a determined country’s sovereign bonds, raising its country risk levels, is the switch of the expected macroeconomic fundamentals’ conditional variance towards a higher regime. Non-linear GARCH models are applied to monitor different switching regimes of the Brazilian country risk conditional volatility, with special emphasis on Markov switching regimes. Results indicate that the high volatility regime periods, better identified by the latter, coincide with all the severe liquidity crisis episodes suffered by Brazil from May 1994 through September 2002. Thus, although not free of limitations, the country risk’s high conditional volatility regime might determine a bad equilibrium and its monitoring might work as a practical tool to assess the duration of liquidity crises in a DSOE highly dependent on foreign capital inflows such as Brazil.