Agricultural Financing And Economic Growth In Nigeria: Short-Run Dynamics And ARDL Bounds Evidence

This study re-examines the relationship between agricultural financing and economic growth in Nigeria using annual data for 1990–2024 and an autoregressive distributed lag (ARDL) bounds-testing framework. The analysis uses gross domestic product as the outcome variable, government expenditure on agriculture and commercial-bank credit to agriculture as financing indicators, and the lending rate as the cost-of-credit control. Augmented Dickey–Fuller and Phillips–Perron tests indicate a mixture of I(0) and I(1) variables and no evidence of I(2) integration. Akaike information criterion selects an ARDL(1,2,2,1) model with a deterministic trend. The bounds F-statistic of 1.6731 falls below the 10% lower bound of 3.0444, so the null hypothesis of no level relationship cannot be rejected. Accordingly, long-run elasticities are not interpreted. In the conditional error-correction representation, the contemporaneous change in government agricultural expenditure has a positive effect on output growth (0.2124; p=0.0097), while its first lag is positive at the 10% level (0.1222; p=0.0855). Changes in bank credit are statistically insignificant, and the contemporaneous lending-rate change is positive but should be treated cautiously because the model shows residual serial correlation and functional-form sensitivity. The findings suggest that agricultural spending can support short-run activity, but the supplied historical series does not establish a stable long run financing–growth relationship. Better-targeted spending, improved credit transmission and longer, consistently measured real series are required for firmer policy inference.