BIS国际清算银行-Understanding-bank-and-non-bank-credit-cycles_-a-structural-exploration_49页_789kb

Summary of "Understanding Bank and Nonbank Credit Cycles: A Structural Exploration"

Core Content

This paper explores the structural drivers of bank and nonbank credit cycles in the United States using a medium-scale Dynamic Stochastic General Equilibrium (DSGE) model. The study highlights the role of sectoral and aggregate financial shocks in influencing credit growth and the broader macroeconomic environment. It also examines the interaction between banks and nonbanks in the financial intermediation system and how their lending activities affect the economy.

Main Points

• Nonbank sector activity has grown significantly over the last three decades, changing the U.S. financial intermediation system. Nearly 60% of credit to the nonfinancial business sector comes from nonbanks.
• The paper investigates structural shocks that drive credit cycles, distinguishing between economy-wide shocks and sector-specific shocks.
• Sectoral shocks are found to be the main drivers of credit growth at the business cycle frequency, with entrepreneur net worth shocks being especially important.
• Aggregate entrepreneurial risk shocks play a significant role in lower-frequency co-movement between bank and nonbank credit cycles.
• Macro shocks have limited impact on financial cycles.
• The model accounts for financial frictions, such as asymmetric information, and leverage dynamics for both banks and nonbanks.
• Banks have capital requirements and deposit insurance, while nonbanks are subject to market discipline and leverage constraints.
• The paper estimates the model using Bayesian methods and U.S. macro and financial data from 1987 to 2015.

Key Findings

• Nonbank lending growth is less volatile than bank lending growth.
• Bank and nonbank lending growth are positively correlated.
• Bank and nonbank lending growth are weakly correlated with investment growth.
• Sectoral shocks (especially entrepreneur net worth shocks) are the primary source of credit growth dynamics.
• Entrepreneur net worth shocks explain around half of the decline in credit growth during the Great Recession.
• Historical decompositions support the model's ability to capture the dynamics of credit cycles.
• The model is externally validated using delinquency rates, corporate bond default rates, and excess bond premium data, which were not used in the estimation process.

Model Structure

• The model includes households, entrepreneurs, investors, and two types of financial intermediaries: banks and nonbanks.
• Households save in bank deposits (riskless) and nonbank deposits (risky).
• Entrepreneurs are divided into two sectors: B-sector (borrowing from banks) and N-sector (borrowing from nonbanks).
• Financial intermediaries (banks and nonbanks) provide liquidity services and intermediate capital to entrepreneurs.
• Default decisions are modeled with limited liability and strategic behavior.
• The model uses a CES function to represent liquidity services derived from deposits and incorporates asymmetric information and leverage constraints.

Empirical Observations

• From 1987 to 2015, the U.S. experienced three distinct credit cycles.
• Bank lending growth declined sharply during the savings and loan crisis, early 2000s recession, and the Great Recession, with nonbank lending showing less severe declines.
• The correlation between bank and nonbank lending growth is positive (0.48), and both are weakly correlated with investment growth (0.16 and 0.17, respectively).
• The model is consistent with these empirical facts, suggesting that it captures the key dynamics of the credit cycle.

Implications

• The study emphasizes the importance of sectoral financial shocks in shaping credit cycles.
• It provides a structural framework to understand how nonbank intermediation interacts with the banking system and affects financial stability.
• The results highlight the need for a more nuanced analysis of credit cycles that accounts for both sectoral and aggregate shocks.

Methodology

• The model is estimated using Bayesian methods.
• Data sources include Federal Reserve Board's Z.1 statistical release and other financial and macroeconomic indicators.
• The model is externally validated using delinquency rates, corporate bond default rates, and excess bond premium data.
• The focus on structural shocks differentiates this paper from reduced-form studies that do not attempt to model the mechanisms behind credit fluctuations.

Conclusion

The paper contributes to the literature by providing a structural model that explains the dynamics of bank and nonbank credit cycles, emphasizing the role of sectoral shocks, especially entrepreneur net worth shocks, in driving credit growth and its implications for the broader economy. The model is robust and consistent with empirical data, offering a comprehensive framework for understanding the interplay between financial intermediation and macroeconomic fluctuations.