Econometric model identifies relationship-level credit shocks
A new ECB working paper proposes a novel econometric model to identify relationship-specific credit demand and supply shocks in bipartite networks. This framework generalizes existing approaches and reveals significant heterogeneity in monetary policy transmission across firm-bank relationships.
Unmasking credit market heterogeneity
The paper introduces a new model for identifying relationship-specific effects in bipartite networks, generalizing the Abowd et al. (1999) framework.
It identifies separate demand and supply shocks for each firm-bank pair, overcoming limitations of previous approaches that assume homogeneity across relationships.
The novel identification strategy yields a consistent and asymptotically normal estimator, requiring weaker network density assumptions.
Applied to thousands of firms and banks across nine Euro-area countries from 2019 to 2023, the methodology formally rejects the standard AKM assumptions in nearly every country-period.
It shows that within-firm/bank shock variation is of comparable scale to between-firm/bank variation, highlighting significant heterogeneity previously masked.
Monetary policy's granular impact
The study documents considerable bias in Abowd et al. (1999) style estimates, where conventional firm-time fixed effects conflate genuine demand with supply factors, leading to counter-intuitive results.
Using the post-2022 ECB monetary contraction as a quasi-natural experiment, the authors find significant deleterious effects on exposed firms, particularly floating-rate borrowers.
These effects on total assets, turnover, and credit-to-assets ratios would be severely underestimated or invisible using conventional methods.
The new framework allows for studying how monetary policy shocks propagate heterogeneously across firm-bank relationships, revealing nuanced impacts.
Sharpening the policy lens
This paper represents a crucial methodological advancement for understanding complex economic relationships.
By disaggregating demand and supply shocks to the relationship level, it provides a far more accurate picture of credit market dynamics and monetary policy transmission.
Policymakers can now better target interventions and assess the heterogeneous impact of their decisions, moving beyond aggregate assumptions.