EFFECTS OF RECAPITALIZATION OF COMMERCIAL BANKS ON AGRICULTURAL FINANCE IN GHANA

Abstract

This study investigates the effects of recapitalization of commercial banks on agricultural finance in Ghana, with emphasis on sectoral loan disbursement dynamics, agricultural credit risk management strategies, and the determinants of credit appraisal scores. Adopting a mixed-methods approach, the study employs both time series and panel data analyses within the framework of Autoregressive Distributed Lag (ARDL) model and fixed effects regression with robust standard errors, respectively, as well as Impulse Response Functions (IRF) and Forecast Error Variance Decomposition (FEVD).The McNemar’s test, cluster analysis and Principal Component Analysis (PCA) were utilized to evaluate shifts in agricultural credit risk management practices and appraisal methodologies following recapitalization, as well as the determinants of credit appraisal scores. The ARDL results revealed a negative and significant adjustment coefficient (-0.942, p < 0.05), confirming stable long-run dynamics between agricultural finance and the independent variables. In the long run, inflation (0.011, p = 0.037) and non-performing loans (NPLs) (0.241, p < 0.05) had a positive and significant impact on agricultural finance. Conversely, capital adequacy ratio (CAR) (-0.034, p = 0.001) and return on equity (ROE) (-0.005, p = 0.042) exhibited negative long-run effects, implying that stringent capital requirements and profitability targets constrain agricultural lending. Recapitalization (RECAP) had a positive but statistically insignificant long-run effect (0.119, p = 0.107), suggesting that increased capitalization does not automatically translate into agricultural credit expansion. This finding aligns with the financial intermediation theory, which posits that although recapitalization enhances the lending capacity of banks by strengthening their capital base, credit allocation decisions remain influenced by sectoral risk perceptions and profitability considerations, thereby explaining the insignificant long-run impact on agricultural finance. The ARDL model's validity was confirmed through rigorous diagnostic tests, ensuring robustness against heteroskedasticity, autocorrelation, endogeneity, and parameter instability. The ECM results further validated these findings, with an error correction term (ECT) of -0.942 (p < 0.05), indicating a rapid and statistically significant adjustment toward long-run equilibrium. In the short run, inflation, NPLs, composite index of economic activities (CIEA), core liquid assets to short-term liabilities (CLSTL), recapitalization, COVID-19 dummy, and monetary policy rate (MPR) had positive and significant effects on agricultural finance. Notably, GDP in the agricultural sector (lnGDPA) (-1.684, p = 0.029) and total assets (lnTA) (-0.03, p < 0.05) had negative short run effects, indicating structural inefficiencies in channeling bank resources to the agricultural sector. The Impulse Response Function (IRF) analysis showed that a shock to recapitalization resulted in a short-lived increase in agricultural finance that dissipated over time, which aligned with the FEVD results, which indicated that recapitalization accounts for only a small but growing share of variance in agricultural finance. The sectoral analysis employing fixed effects regression demonstrated the heterogeneous impact of recapitalization across economic sectors. While sectors such as services, commerce, and transportation attracted significantly higher loan disbursements relative to agriculture, the mining sector was the only sector with an inverse relationship. This finding confirms a sectoral bias in credit allocation, which necessitates targeted interventions to redirect credit flows toward agriculture. The McNemar’s test revealed a strategic shift in agricultural credit risk management strategies, with banks transitioning from traditional collateral based lending to data-driven, structured risk management approaches, including specialized agricultural credit scoring models. The cluster analysis identified three distinct lender profiles which are collateral-oriented lenders, balanced financial-statement-based lenders, and inclusive development-oriented lenders, with each adopting unique credit appraisal methodologies. Furthermore, Principal Component Analysis (PCA) identified five critical components influencing agricultural credit appraisal scores. These are risk based factors, regulatory compliance, operational efficiency and technology adoption, market resilience, and collateral strength. These determinants reflect the multifaceted nature of agricultural credit assessment and justify the need for tailored credit models that account for both financial and non-financial borrower characteristics. The findings of the study affirm recapitalization as a secondary driver of agricultural credit, with macroeconomic stability and credit quality serving as primary determinants. This study contributes to literature by providing empirical evidence on the interactions between recapitalization, sectoral loan distribution, and agricultural financing or credit in emerging markets. The findings affirm the need for policy frameworks that not only strengthen financial institutions but also incentivize agricultural lending, ensuring that recapitalization reforms translate into sustained growth within the agricultural sector.