Impact of Listed Firms’ Correlation on Idiosyncratic Volatility Co-movement—A Network and Wavelet Analysis

Idiosyncratic risk reflects the specific and non-systemic risk of a firm. Based on Capital Asset Pricing Model (CAPM), idiosyncratic volatility (IVOL) measures a portion of the variation in asset returns that cannot be explained by a particular CAPM. Existing literature shows that there is a co-movement phenomenon among listed-firms, which contradicts the definition of IVOL. Analogous to the inter-stock correlation phenomenon, we test whether the inter-firm network relationship is the underlying reason for the IVOL co-movement phenomenon among the listed-firms. Using a comparative analysis method, we construct three network correlation structures (geographic distance, stock return correlation, and risk transmission) and examine whether the degree of IVOL co-movement significantly decreases after eliminating various network effects. Wavelet analysis reveals that the co-movement phenomenon persists even after accounting for network correlations, suggesting the presence of unique relationships among the firms, such as interpersonal relationship (Chinese ‘Guanxi’ culture). Given the results, we suggest that policy makers can focus on the unobserved interpersonal relationship network when determining the portfolios.