Please rotate your device to landscape mode to view the charts.

Background and Context

Research Focus

This study examines how financial markets and institutions affect bank performance across 93 countries from 2008-2023.

ESG Integration

Environmental, Social and Governance readiness is analyzed as both a direct driver and moderator of banking outcomes.

Analytical Approach

Panel quantile regression captures nonlinear effects and handles non-normally distributed data with outliers across countries.

Financial Institutions Show U-Shaped Impact on Bank Performance

Initial Decline Extended Recovery Financial Institution Development Bank Performance Turning Point
  • The relationship between financial institution development and bank performance follows a U-shaped pattern over time.
  • Initially, financial development reduces bank performance due to increased competition and market saturation pressures.
  • Beyond a threshold, continued financial development improves bank soundness through innovation and efficiency gains.

Financial Institution Effects: Strong U-Shape with Significant ESG Moderation

  • The negative FIP coefficient (-16.84) confirms initial adverse effects of financial institution development on banks.
  • The positive FIP² coefficient (20.75) validates the U-shaped recovery as financial development continues further.
  • ESG moderation effects (FIP×ESG: 28.89, FIP²×ESG: -33.66) flatten the curve and accelerate positive outcomes.

Entrepreneurship Boosts Bank Performance While Internet Usage Reduces It

  • Entrepreneurship consistently increases bank performance (0.03) by diversifying loan portfolios and reducing risk concentration.
  • Internet usage negatively affects bank soundness (-0.01) due to cybersecurity costs and unregulated fintech competition.
  • Mobile usage shows small negative effects, though not statistically significant in the median quantile estimates.

ESG Readiness Transforms and Stabilizes the Financial Development-Performance Link

Low ESG High ESG Financial Development Bank Performance ESG Lifts & Flattens Curve
  • High ESG readiness shifts the entire relationship upward, meaning banks perform better at all development levels.
  • ESG flattens the U-curve, reducing the initial negative phase and accelerating the transition to positive outcomes.
  • This moderating effect confirms ESG promotes stability and reduces vulnerability to financial system shocks.

Financial Institutions Model Explains More Variance Than Financial Markets Model

  • Both models explain over 60% of variance in bank performance, indicating strong explanatory power for the framework.
  • The Financial Institutions model slightly outperforms, suggesting institutional factors matter more than market factors.
  • Robust M-estimation confirms results hold under alternative specifications, with R² reaching 67% for the FIP model.

Contribution and Implications

  • First study to simultaneously examine financial markets and institutions' effects on bank performance within unified framework.
  • Demonstrates ESG readiness can accelerate positive outcomes and reduce transition time from negative to positive effects.
  • Provides evidence that policymakers should embed ESG compliance benchmarks as standard practice within banking regulation.
  • Suggests central banks should integrate environmental targets with monetary policy through preferential green financing rates.
  • Highlights need for enhanced cybersecurity infrastructure as internet banking expansion can negatively impact bank soundness.

Data Sources

  • Finding 1 (U-shaped relationship): Conceptual visualization based on the theoretical model and significant coefficients from Table 4.
  • Finding 2 (FIP coefficients): Bar chart constructed using regression coefficients from Table 4 FIP Model (FIP=-16.84, FIP²=20.75, FIP×ESG=28.89, FIP²×ESG=-33.66).
  • Finding 3 (Control variables): Chart uses coefficient values from Table 4 for both models (ENT=0.03, INT=-0.01, MOB=-0.001/-0.002).
  • Finding 4 (ESG moderation): SVG visualization based on Figures 6 and 7 in the article showing moderating effect patterns.
  • Finding 5 (Model comparison): R² values from Table 4 (0.61, 0.62) and Table 6 (0.65, 0.67) for quantile and M-estimation approaches.