The Cross‐Industry Contagion Network of Systemic Risk: Evidence From China

ABSTRACT This study employs the Lasso–Granger algorithm, in combination with sliding window technology, to construct a dynamic directed complex network that captures the dynamics of systemic risk transmission between the financial industry and the real economy. By analysing the network's characteristic indicators among the banking, securities, and insurance sectors, we explore variations in systemic risk transmission across these three major financial sectors and the real economy. Additionally, the CONCOR algorithm is used to detect community structures within the complex network, enabling a dynamic assessment of these communities' attributes over time. The empirical results show that the ability of the insurance, banking, and securities sectors to receive systemic risk weakens progressively. The insurance sector has the strongest capacity to transmit systemic risk, whereas the banking sector has the weakest capacity to transmit systemic risk. The banking and insurance sectors exhibit similar role attributes in the process of systemic risk transmission. The risk transmission relationship between the real economy and the financial industry shows heterogeneity.