Dependence structure and dynamic connectedness between green bonds and financial markets: fresh insights from time-frequency analysis before and during COVID-19 pandemic

Abstract

This paper examines the interdependence between green bonds and financial markets in the timefrequency domain by utilizing the multivariate wavelet approach and dynamic connectedness through combining Ensemble Empirical Mode Decomposition (EEMD) with Diebold and Yilmaz (2012) spillover framework. The findings of wavelet multiple correlations indicate that the benefits of diversification opportunities are more evident in the short run. The evidence of wavelet multiple crosscorrelations reveals that green bonds and financial markets are highly integrated in the long run. The results of the static connectedness framework explain that the direction and magnitude of spillover behave differently across markets. The world stock market is the net spillover transmitter, while the corporate bond market is the net spillover receiver among the selected markets. The green bond market is receiving more but transmitted less volatility in the present study. The evidence on dynamic connectedness measured by the rolling window approach shows that the interconnection between green bonds and financial markets is volatile over time. These pieces of evidence provide implications to global investors having a strong position in the green bonds market in terms of risk management and portfolio decisions.