Monetary policy shocks and the dynamic response of bank risk: evidence of nonlinear transmission through interest rate exposure

This paper investigates the dynamic and state-dependent transmission of monetary policy shocks to bank risk, focusing on the role of balance sheet interest rate exposure. Using high-frequency identification to construct exogenous monetary policy shocks for China, we combine local projections with local polynomial regression to analyse a comprehensive panel of 229 commercial banks. Results from local projections indicate that contractionary monetary policy shocks significantly increase bank risk, following a hump-shaped pattern over a five-year horizon: an immediate, substantial impact, followed by attenuation and a resurgence in later horizons. The shock effects are asymmetric, being more pronounced under tightening shocks. Local polynomial regression further reveals that the transmission mechanism exhibits highly nonlinear characteristics, reflecting a trade-off between marginal gains and marginal losses. Macroprudential policies are found to partially mitigate these risks. Our findings, robust to extensive endogeneity checks, suggest that balance sheet composition is a critical determinant of monetary policy’s impact on financial stability, offering key insights for risk management and regulatory design in emerging markets.