Purpose This study examines the lagged effects of macroeconomic and bank-specific variables on agricultural non-performing loans (NPLs) in Ghana, addressing a gap in modelling delayed transmission channels within a sector characterised by long production cycles and structural financing constraints. Design/methodology/approach Using monthly banking data from January 2015 to December 2022, the study employs a two-stage strategy: an autoregressive distributed lag (ARDL) bounds test to confirm cointegration, followed by Prais–Winsten feasible generalised least squares with AR(1) correction as the primary inferential model. A binary dummy captures the structural shift associated with Ghana's 2017–2019 banking recapitalisation. Findings Agricultural GDP growth is inversely associated with NPLs in the short run (β = −0.237) and long run (β = −0.482), consistent with the financial accelerator framework. Inflation and the monetary policy rate exert significant positive effects through cost-push and debt-service channels, respectively. Capital adequacy is inversely related to NPLs, while recapitalisation is the largest short-run determinant (β = −0.091), operating through institutional consolidation and regulatory signalling. The error correction term (−0.312) confirms long-run equilibrium adjustment within approximately 3.2 months. Practical implications Results support sustained prudential capital standards, sector-disaggregated NPL monitoring and countercyclical instruments such as GIRSAL to mitigate adverse effects of monetary tightening on agricultural credit quality. Originality/value This paper is among the first to systematically model lagged agricultural NPL determinants in Ghana, demonstrating that recapitalisation's structural effects substantially exceed the continuous capitalisation channel, with actionable implications for regulators and agricultural lenders in frontier markets.
Akwasi Agyeman Britwum (Tue,) studied this question.