The Interplay of CBN Liquidity Provision, Incentive-Driven Risk Behavior, and Market-Institutional Perceptions in Monetary Policy Transmission

Vollständiger Abstract

Worum geht es in dieser Arbeit?

This study investigates the interplay of central-bank liquidity provision, incentive-driven risk behavior, and market-institutional perceptions in the transmission of monetary policy in Nigeria, with a specific focus on the discount-window facility of the Central Bank of Nigeria (CBN). The research is motivated by the ongoing debate about whether the CBN’s liquidity operations merely serve as a safety net or whether they can be leveraged to influence bank profitability and risk-taking, especially in the context of the 2024 recapitalization requirement that mandates Nigerian banks to raise their minimum paid-up capital to ₦50 billion. The empirical analysis draws on a balanced quarterly panel of fifteen Nigerian banks spanning 2010Q1 to 2023Q4, yielding 56 observations per bank. The Pedroni test fails to reject the null hypothesis of no cointegration across all seven test statistics (p > 0.10), indicating that the series do not share a common long-run equilibrium. Given the absence of cointegration, a dynamic short-run model is estimated using the one-step Arellano-Bond generalized method of moments (GMM) estimator, which accounts for potential endogeneity of discount-window variables and exploits the panel’s lag structure. The GMM specification includes lagged profitability, lagged discount-window volume, lagged discount-window cost, the recapitalization dummy, the interaction term, and the control variables. Robust standard errors are employed, and instrument validity is confirmed by a J-statistic of 28.14 (p = 0.614) and the absence of first- and second-order serial correlation in the differenced residuals (AR(1) = -1.42, p = 0.156; AR(2) = 0.78, p = 0.435). The GMM results reveal that the coefficient on lagged discount-window volume is 0.007 (t = 0.39, p = 0.697), suggesting a statistically insignificant impact on profitability, thereby supporting H1. Likewise, the coefficient on lagged discount-window cost is –0.002 (t = -0.50, p = 0.618), indicating that the price of discount-window funds does not materially alter banks’ risk profiles, which confirms H2. The recapitalization dummy and its interaction with borrowing volume also yield negligible coefficients (0.003, p = 0.618 and 0.001, p = 0.912, respectively), providing evidence that the 2024 capital requirement does not moderate the relationship between discount-window borrowing and bank performance, thus validating H3. Robustness checks—including alternative lag structures, expanded instrument sets, and a system-GMM estimator—produce qualitatively identical findings, affirming the reliability of the results. The study concludes that, within the examined sample, the CBN’s discount-window facility functions primarily as a liquidity backstop rather than a tool for influencing bank profitability or risk-taking.