Monitoring Financial Stress in South Africa

This article develops a new index to monitor financial stability in South Africa over the period 1995-2017. Rather than selecting indicators based on deemed relevance, the novelty of our index lies in the selection and aggregation of financial indicators based on their incremental informational content, achieving the best balance between parsimony and efficacy. In addition, market sub-indices are weighted by time-varying cross-correlations among them, enabling the financial stress measure to focus on the systemic dimension of financial stress. While capturing the key episodes of financial stress in South Africa, the index also successfully captures other global and idiosyncratic risks that affect the financial markets in the country. Threshold vector autoregression models based on the full sample and sub-samples reveal nonlinearities and time-variation in the transmission of a financial shock to the real economy.

Automated summary

This study develops a new index to monitor financial stability in South Africa, selecting and aggregating financial indicators based on their incremental informational content to achieve the best balance between parsimony and efficacy. The index successfully captures not only key episodes of financial stress in South Africa, but also other global and idiosyncratic risks affecting the country's financial markets. Threshold vector autoregression models reveal nonlinearities and time-variation in the transmission of a financial shock to the real economy.