Sectoral integration on an emerging stock market: a multi-scale approach

Abstract

The purpose of this study is to examine the connectedness of industry sectors on the Johannesburg Stock Exchange in a time–frequency domain. We use econophysics-based methods like the wavelet multiple correlation and wavelet scalogram difference to identify the evolution of the connectedness of the sectors over time and at different frequencies. The findings show that the sectors on the Johannesburg Stock Exchange are especially integrated at lower frequencies. Wavelet multiple correlation peaks in response to local and global shocks like the black-swan COVID-19 pandemic in 2020 and the downgrading of South African debt by Fitch in 2013. Though there are opportunities for sectoral diversification on the JSE, this fails when it is most needed, during crisis periods. Investors should therefore consider other asset classes that could serve as a haven in times of crisis. Though extant literature has examined sectoral dependencies on the stock markets of developed and developing countries, to the best of our knowledge, this is the first study to examine this connectedness in a South African context using multiple nonparametric methods that are robust to non-normality, presence of outliers as well as non-stationary data.