国际清算银行-长期债务传播和实际逆转（英）-2023.5-54页_673kb

Long-Term Debt Propagation and Real Reversals

Summary

This paper by Drehmann, Juselius, and Korinek introduces a new mechanism for how credit cycles propagate through the economy via long-term debt, leading to predictable real reversals.

Key findings:

1. New borrowing vs. Debt service: While new borrowing boosts output contemporaneously, debt service (future repayments) depresses output, leading to predictable medium-term economic slowdowns.

2. Lead-lag relationship: The study documents a significant time lag of 4-7 years between peaks in new household borrowing and peaks in debt service payments, mainly due to amortization schedules and correlated new borrowing patterns.

3. Data contribution: Provides the first systematic cross-country database (16 advanced economies, 1980-2019) of household debt flows, highlighting the importance of tracking debt service and new borrowing flows for financial-cycle analysis.

4. Economic significance:

   • A 1 percentage point increase in new borrowing raises 1-year-ahead GDP growth by approximately 12 basis points
   • A 1 percentage point increase in debt service reduces 1-year-ahead GDP growth by approximately 19 basis points

5. Policy implications: Economic policymakers face a trade-off between short-term stimulus through borrowing and long-term drag from debt service. This mechanism highlights why monetary policy may struggle during debt repayment periods even with low interest rates.

6. Measurement focus: Reinforces the need to better measure financial flows (both maturities and repayment patterns) between borrowers and lenders, especially for the corporate sector.

The paper contributes a simple but powerful framework for understanding the pervasive link between financial cycles and real economic activity, emphasizing the crucial role of debt service flows. This mechanism largely accounts for the predictive power of credit cycle measures observed in earlier research.