Regional disparities in the implementation of farmer–producer organizations: a machine learning approach

Purpose A study was conducted to analyse the extent of farmer–producer organization (FPO) implementation and its determinants in Karnataka. Districts were classified into High and Low implementation groups based on the number of FPOs per gross cropped area (’000 ha), with 81.24% of high-implementation districts in Southern Karnataka and only 14.29% in the North, revealing a significant regional disparity. Design/methodology/approach To identify key factors of this variation, support vector machine (SVM), a supervised machine learning technique, was employed. Findings Promoting institutions (PIs) (+0.81), primary agricultural credit societies (PACS) (+0.63) and Regulated Markets (+0.19) are the most influential drivers classifying districts as high in FPO implementation, emphasizing the importance of institutional facilitation, credit and input access and market linkages. Weights of average landholding (−1.22), farm income per farmer (−0.37) and cropping intensity (−0.39) indicate that districts with more small and marginal farmers, lower income and lesser cropping intensity, which are the typical features of the south, have higher FPO implementation. The disparity is observed in Northern districts, which are predominantly agrarian with larger landholdings, limited non-farm income and reduced access to markets and credit. Results of Ridge logit also revealed that a one-unit increase in PIs raises the odds of high FPO implementation by 160.2% (OR = 2.602). Markets and PACS also showed significant effects with odds ratios of 1.884 (88.4%) and 1.470 (47.0%), respectively. Research limitations/implications It is finally concluded that expanding region-specific high cropping intensity and income potential districts of Northern Karnataka through PIs markets and credit access may further strengthen the agrarian economy and improve rural livelihoods. Originality/value This study integrates SVM feature weights with ridge logistic regression to quantify drivers of regional disparity in FPO implementation, offering evidence for targeted, region-specific interventions in Karnataka.