On the Relationship between Financial Instability and Economic Performance: Stressing the Business of Nonlinear Modelling

The recent global financial crisis and the subsequent economic recession have revitalized the discussion on the causal relationship between financial and economic sectors. This study examines financial and economic indices developed by the Federal Reserve Banks of Kansas City and Chicago, respectively, to identify the impact of financial uncertainty on the overall economic performance. Using smooth transition and vector smooth transition autoregressions, this research assesses nonlinear dynamics of these indices, and tests the Granger non-causality hypothesis between the financial stress and economic activity in an out-of-sample setting. Results of this study confirm the causal relationship between financial and economic indices. Moreover, the improved in-sample fit of nonlinear models translates into better forecast performance, in comparison with the linear models, in an out-of-sample setting as well.