Factor models and mutually exciting jump processes for financial systemic risk

We provide a reduced-form model for the propagation of idiosyncratic shocks from any specific economic unit to the entire financial system. Using common factors and mutually exciting jumps in both price and volatility, we distinguish between sources of systemic failure such as macro risk drivers, connectedness, and contagion. The estimation procedure relies on the Generalized Method of Moments and takes advantage of high-frequency data. We use model parameters to define networks for shock transmission, and we provide new measures for financial system fragility. We construct paths for shock propagation within key US financial institutions and within S&P500 sectors, over the period 2006–2024. We find that beyond common factors, systemic dependency has two distinct channels, price and volatility jumps, with volatility-driven contagion exhibiting stronger and more pervasive effects. The weighted connectivity and spectral indicators reveal that larger, more central institutions amplify system-wide disturbances, while sectoral analysis shows that contagion extends beyond finance into production-oriented industries. The findings indicate that volatility spillovers dominate price-based contagion, underscoring the need for regulatory policies that jointly address interconnectedness and uncertainty-driven transmission mechanisms.