This paper develops a credit scoring model for agricultural loan portfolio of a large public sector bank in India and suggests how such a model would help the bank to mitigate risk in agricultural lending. The logistic model developed in this study captures the major risk drivers in agricultural loan portfolio and designed to be consistent with Basel II, including consideration given to forecasting accuracy and model applicability. The significant risk factors are facility type, cropping pattern, borrower characters, cost of living, regional locations and collateral/security type beside borrowers' financial capacity and leverage position. The study also shows how agricultural exposures can be typically managed on a portfolio basis which will not only enable the bank to diversify the risk and optimize the profit in the business, but also will strengthen banker-borrower relationship and enables the bank to expand its reach to farmers because of transparency in loan decision making process.

A rapid growth in the rural economy and within that of agriculture in India is highly feasible provided key ingredients such as adequate supply of credit and the availability of the tools for the management of risks that agriculture is exposed to are religiously followed.

Farm level surveys have indicated that the most frequently cited risks are price, crop/weather and health. These risks among others could lower farmers' anticipated income and have negative effects on their standard of living, ability to provide for themselves and their families, ability to build capital and hence, their inherent creditworthiness. In order to sustain credit disbursement to agricultural farmers, public sector banks in India should be able to ease risk arising from credit exposure in agriculture. A good credit risk assessment assists banks and financial institutions in taking right and informed credit decisions, proper loan pricing, determining the amount of loans to be disbursed, reducing the chance of default and finally, increasing the possibility of debt recovery. Credit risk assessment involves determining the financial strength of the borrowers, estimating the probability of default and reducing the risk of non-payment to an acceptable level. In general, credit evaluations in public sector banks in India are based on the credit officer's subjective assessment of judgmental assessment techniques.

However, this technique seems to be inefficient, inconsistent and above all non-uniform because of subjectivity in choice of risk weights and scores, and hence, suboptimal. Rather, customized credit scoring model based on internal data of a bank has the potential of reducing the variability of credit decisions and imparting efficiencies to credit risk assessment process.

Credit Risk Models, Bank's Agricultural Loan Portfolio, Logistic model, Cropping patterns, Statistical model approach, Agricultural sector, Financial capacity, Judgmental assessment techniques, Public sector banks, Decision making process, Credit risk assessment, Non-Performing Assets.