2015年-世界发展银行全球_How_Bank_Competition_Affects_Firms_Access_to_Finance_36页_256kb

Summary of "How Bank Competition Affects Firms' Access to Finance"

Core Content

This paper investigates the relationship between bank competition and firms' access to finance using firm-level survey data from 53 countries and a panel dataset covering nearly 70,000 firms. The study aims to resolve the ambiguity in existing theories regarding the impact of competition on financial access.

Main Theories and Hypotheses

Two primary hypotheses are discussed:

1. Market Power Hypothesis: Suggests that low competition (high market power) reduces firms' access to finance by increasing the cost and decreasing the availability of credit.
2. Information Hypothesis: Proposes that low competition may actually improve access to finance because banks can internalize the costs of building firm-specific relationships, especially with opaque clients.

The paper rejects the information hypothesis and supports the market power hypothesis through empirical evidence.

Key Findings

• Low bank competition (measured by high values on the Lerner index or Boone indicator) is associated with diminished access to finance for firms.
• Credit information sharing mechanisms play a crucial role in mitigating the negative impact of low competition. Better credit information systems reduce the adverse effects of low competition on firms’ ability to access finance.
• Private credit bureaus help reduce the negative impact of low competition, whereas public credit registries do not.
• The effect of competition on access to finance is more pronounced in environments with higher initial costs of finance, suggesting that credit information sharing lowers these costs and thus reduces the importance of competition.
• Capital markets as an alternative source of financing are only weakly supportive of the hypothesis that they can counteract the negative effects of bank competition.

Methodology and Data

• The study uses firm-level data from the World Bank Enterprise Surveys, bank-level data from Bankscope, and country-level data to construct variables for analysis.
• Two main datasets are used:
  • A large cross-sectional dataset of over 68,000 firms from 53 countries for the period 2002–2010.
  • A smaller panel dataset of approximately 14,000 firms from 42 countries with multiple observations per firm.
• Non-structural measures of competition, such as the Lerner index and Boone indicator, are used instead of traditional concentration measures, which are found to be not significant.
• The Lerner index measures the degree to which banks can charge prices above marginal cost.
• The Boone indicator reflects profit elasticity and is interpreted as a measure of competition.
• The paper also includes concentration measures (Concentration 3 and Herfindahl index) to test the structure-conduct-performance paradigm.

Contributions

• The study extends previous research by:
  • Using a larger international dataset and panel data techniques to control for country-specific fixed effects.
  • Employing an objective measure of firm access to finance (whether a firm uses credit products) rather than subjective or proxy measures.
  • Providing a direct test of the interaction between competition and credit information sharing mechanisms.
  • Examining the role of private vs. public credit information systems in mitigating the negative effects of low competition.

Conclusion

The paper concludes that low bank competition reduces firms' access to finance, and this effect is lessened by the presence of effective credit information sharing mechanisms, particularly private ones. These findings support the market power hypothesis and suggest that credit information sharing can serve as a substitute for bank-firm relationships, improving access to finance in less competitive banking environments.

Key Variables and Measures

Additional Notes

• The study controls for firm-specific characteristics such as size, industry (manufacturing vs. service), export status, foreign ownership, and firm age.
• The objective measure of access to finance is preferred over subjective or proxy measures to ensure comparability across countries and reliability.
• The interaction analysis between competition and credit information sharing provides more direct evidence for the market power hypothesis.